Groover fundamentals modern manufacturing 4th txtbk 11

E1C38 11/09/2009 18:1:1 Page 886

Part X ManufacturingSystems

38AUTOMATIONTECHNOLOGIES FORMANUFACTURINGSYSTEMS

Chapter Contents

38.1 Automation Fundamentals38.1.1 Three Components of an Automated

System38.1.2 Types of Automation

38.2 Hardware Components for Automation38.2.1 Sensors38.2.2 Actuators38.2.3 Interface Devices38.2.4 Process Controllers

38.3 Numerical Control38.3.1 The Technology of Numerical Control38.3.2 Analysis of NC Positioning Systems38.3.3 NC Part Programming38.3.4 Applications of Numerical Control

38.4 Industrial Robotics38.4.1 Robot Anatomy38.4.2 Control Systems and Robot

Programming38.4.3 Applications of Industrial Robots

In this part of the book, we consider the manufacturingsystems that are commonly associated with the productionand assembly processes discussed in preceding chapters. Amanufacturing system can be defined as a collection of inte-grated equipment and human resources that performs one ormore processing and/or assembly operations on a startingwork material, part, or set of parts. The integrated equipmentconsists of production machines, material handling and posi-tioning devices, and computer systems. Human resources arerequired either full-time or part-time to keep the equipmentoperating. The position of the manufacturing systems in thelarger production system is shown in Figure 38.1. As thediagram indicates, the manufacturing systems are located inthe factory. They accomplish the value-addedwork on the partor product.

Manufacturing systems include both automated andmanually operated systems. The distinction between the twocategories is not always clear, because many manufacturingsystems consist of both automated and manual work ele-ments (e.g., amachine tool that operates on a semiautomaticprocessing cycle but which must be loaded and unloadedeach cycle by a human worker). Our coverage includes bothcategories and is organized into two chapters: Chapter 38 onautomation technologies and Chapter 39 on integratedmanufacturing systems.Chapter 38 provides an introductory

886

E1C38 11/09/2009 18:1:1 Page 887

treatment of automation technology and the components that make up an automatedsystem. We also discuss two important automation technologies used in manufacturing:numerical control and industrial robotics. InChapter 39,weexaminehow theseautomationtechnologies are integrated intomore sophisticatedmanufacturing systems. Topics includeproduction lines, cellular manufacturing, flexible manufacturing systems, and computerintegrated manufacturing. A more detailed discussion of the topics in these two chapterscan be found in [5].

38.1 AUTOMATION FUNDAMENTALS

Automation can be defined as the technology by which a process or procedure isperformed without human assistance. Humans may be present as observers or evenparticipants, but the process itself operates under its own self-direction. Automationis implemented by means of a control system that executes a program of instructions.To automate a process, power is required to operate the control system and to drivethe process itself.

38.1.1 THREE COMPONENTS OF AN AUTOMATED SYSTEM

As indicated above, an automated system consists of three basic components: (1) power,(2) a program of instructions, and (3) a control system to carry out the instructions. Therelationship among these components is shown in Figure 38.2.

The form of power used in most automated systems is electrical. The advantages ofelectrical power include (1) it is widely available, (2) it can be readily converted to otherforms of power such as mechanical, thermal, or hydraulic, (3) it can be used at very lowpower levels for functions such as signal processing, communication, data storage, anddata processing, and (4) it can be stored in long-life batteries [5].

FIGURE 38.1 Theposition of the manufac-turing systems in thelarger production system.

Manufacturing processes and assembly operations

Facilities

Manufacturingsupport

Quality controlsystem

Manufacturingsystems

Manufacturingsupport systems

Production system

Finishedproducts

Engineeringmaterials

FIGURE 38.2 Elements

of an automated system:(1) power, (2) programof instructions, and (3)control system.

(1) Power

(2) Program ofinstructions

(3) Controlsystem

ProcessProcess output

Section 38.1/Automation Fundamentals 887

E1C38 11/09/2009 18:1:1 Page 888

In amanufacturing process, power is required to accomplish the activities associatedwith the particular process. Examples of these activities include (1) melting a metal in acasting operation, (2) driving the motions of a cutting tool relative to a workpiece in amachining operation, and (3) pressing and sintering parts in a powder metallurgy process.Power is also used to accomplish any material handling activities needed in the process,such as loading and unloading parts, if these activities are not performedmanually. Finally,power is used to operate the control system.

The activities in an automated process are determined by a program of instructions. Inthe simplest automated processes, the only instruction may be to maintain a certaincontrolled variable at a specified level, such as regulating the temperature in a heat treatmentfurnace. In more complex processes, a sequence of activities is required during the workcycle, and the order and details of each activity are defined by the program of instructions.Each activity involves changes in one or more process parameters, such as changing the x-coordinate position of a machine tool worktable, opening or closing a valve in a fluid flowsystem, or turning a motor on or off. Process parameters are inputs to the process. They maybe continuous (continuously variable over a given range, such as the x-position of aworktable) or discrete (On or Off). Their values affect the outputs of the process, whichare called process variables. Like process parameters, process variables can be continuousor discrete. Examples include the actual position of the machine worktable, the rotationalspeed of a motor shaft, or whether a warning light is on or off. The program of instructionsspecifies the changes in process parameters and when they should occur during the workcycle, and these changes determine the resulting values of the process variables. For example,in computer numerical control, the program of instructions is called a part program. Thenumerical control (NC) part program specifies the individual sequence of steps required tomachine a given part, including worktable and cutter positions, cutting speeds, feeds, andother details of the operation.

In some automated processes, the work cycle programmust contain instructions formaking decisions or reacting to unexpected events during the work cycle. Examples ofsituations requiring this kind of capability include (1) variations in raw materials thatrequire adjusting certain process parameters to compensate, (2) interactions and com-munications with human such as responding to requests for system status information,(3) safety monitoring requirements, and (4) equipment malfunctions.

The program of instructions is executed by a control system, the third basiccomponent of an automated system. Two types of control system can be distinguished:closed loop and open loop. A closed loop system, also known as a feedback control system,is one in which the process variable of interest (output of the process) is compared with thecorresponding process parameter (input to the process), and any difference between themis used to drive the output value into agreement with the input. Figure 38.3(a) shows the sixelements of a closed loop system: (1) input parameter, (2) process, (3) output variable, (4)feedback sensor, (5) controller, and (6) actuator. The input parameter represents thedesired value of the output variable. The process is the operation or activity beingcontrolled; more specifically, the output variable is being controlled by the system. Asensor is used to measure the output variable and feed back its value to the controller,which compares output with input and makes the required adjustment to reduce anydifference. The adjustment is made by means of one or more actuators, which arehardware devices that physically accomplish the control actions.

The other type of control system is an open loop system, presented in Figure 38.3(b).As shown in the diagram, an open loop system executes the program of instructionswithout a feedback loop. No measurement of the output variable is made, so there is nocomparison between output and input in an open loop system. In effect, the controllerrelies on the expectation that the actuator will have the intended effect on the outputvariable. Thus, there is always a risk in an open loop system that the actuator will notfunction properly or that its actuation will not have the expected effect on the output. On

888 Chapter 38/Automation Technologies for Manufacturing Systems

E1C38 11/09/2009 18:1:2 Page 889

the other hand, the advantage of an open loop system is that its cost is less than acomparable closed loop system.

38.1.2 TYPES OF AUTOMATION

Automated systems used in manufacturing can be classified into three basic types:(1) fixed automation, (2) programmable automation, and (3) flexible automation.

Fixed Automation In fixed automation, the processing or assembly steps and theirsequence are fixed by the equipment configuration. The program of instructions is deter-mined by the equipment design and cannot be easily changed. Each step in the sequenceusually involves a simple action, such as feeding a rotating spindle along a linear trajectory.Although the work cycle consists of simple operations, integrating and coordinating theactions can result in the need for a rather sophisticated control system, and computer controlis often required.

Typical features of fixed automation include (1) high initial investment for specializedequipment, (2) high production rates, and (3) little or no flexibility to accommodate productvariety. Automated systems with these features can be justified for parts and products thatare produced in very large quantities. The high investment cost can be spread over manyunits, thus making the cost per unit relatively low compared to alternative productionmethods. The automated production lines discussed in the following chapter are examplesof fixed automation.

Programmable Automation As its name suggests, the equipment in programmableautomation is designed with the capability to change the program of instructions to allowproduction of different parts or products. New programs can be prepared for new parts, andthe equipment can read each program and execute the encoded instructions. Thus thefeatures that characterize programmable automation are (1) high investment in generalpurpose equipment that can be reprogrammed, (2) lower production rates than fixedautomation, (3) ability to cope with product variety by reprogramming the equipment, and(4) suitability for batch production of various part or product styles. Examples of program-mable automation include computer numerical control and industrial robotics, discussed inSections 38.3 and 38.4, respectively.

Flexible Automation Suitability for batch production is mentioned as one of thefeatures of programmable automation. As discussed in Chapter 1, the disadvantage

FIGURE 38.3 Two basictypes of control systems:(a) closed loop and

(b) open loop.

Controller

(1)Input

parameter

(3)Outputvariable

Inputparameter

(5) (6)

(4)

(a)

(b)

(2)

Actuator

Feedbacksensor

Process

Controller Actuator ProcessOutputvariable

Section 38.1/Automation Fundamentals 889

E1C38 11/09/2009 18:1:2 Page 890

of batch production is that lost production time occurs between batches due toequipment and/or tooling changeovers that are required to accommodate the nextbatch. Thus, programmable automation usually suffers from this disadvantage.Flexible automation is an extension of programmable automation in which thereis virtually no lost production time for setup changes and/or reprogramming. Anyrequired changes in the program of instructions and/or setup can be accomplishedquickly; that is, within the time needed to move the next work unit into position at themachine. A flexible system is therefore capable of producing a mixture of differentparts or products one right after the other instead of in batches. Features usuallyassociated with flexible automation include (1) high investment cost for custom-engineered equipment, (2) medium production rates, and (3) continuous productionof different part or product styles.

Using some terminology developed in Chapter 1, we might say that fixed automa-tion is applicable in situations of hard product variety, programmable automation isapplicable to medium product variety, and flexible automation can be used for softproduct variety.

38.2 HARDWARE COMPONENTS FOR AUTOMATION

Automation and process control are implemented using various hardware devices thatinteract with the production operation and associated processing equipment. Sensors arerequired to measure the process variables. Actuators are used to drive the processparameters. And various additional devices are needed to interface the sensors andactuators with the process controller, which is usually a digital computer.

38.2.1 SENSORS

A sensor is a device that converts a physical stimulus or variable of interest (e.g.,temperature, force, pressure, or other characteristic of the process) into a more convenientphysical form (e.g., electrical voltage) for the purpose of measuring the variable. Theconversion allows the variable to be interpreted as a quantitative value.

Sensors of various types are available to collect data for feedback control inmanufacturing automation. They are often classified according to type of stimulus;thus, we have mechanical, electrical, thermal, radiation, magnetic, and chemical variables.Within each category, there are multiple variables that can be measured. Within themechanical category, the physical variables include position, velocity, force, torque, andmany others. Electrical variables include voltage, current, and resistance. And so on forthe other major categories.

In addition to type of stimulus, sensors are also classified as analog or discrete. Ananalog sensor measures a continuous analog variable and converts it into a continuoussignal such as electrical voltage. Thermocouples, strain gages, and ammeters are exam-ples of analog sensors. A discrete sensor produces a signal that can have only a limitednumber of values. Within this category, we have binary sensors and digital sensors. Abinary sensor can take on only two possible values, such as Off and On, or 0 and 1. Limitswitches operate this way. A digital sensor produces a digital output signal, either in theform of parallel status bits, such as a photoelectric sensor array) or a series of pulses thatcan be counted, such as an optical encoder. Digital sensors have an advantage that they canbe readily interfaced to a digital computer, whereas the signals from analog sensors mustbe converted to digital in order to be read by the computer.


E1C38 11/09/2009 18:1:2 Page 891

For a given sensor, there is a relationship between the value of the physical stimulusand the value of the signal produced by the sensor. This input/output relationship is calledthe sensor’s transfer function, which can be expressed as:

S ¼ f sð Þ ð38:1Þ

where S ¼ the output signal of the sensor (typically voltage), s ¼ the stimulus or input,and f(s) is the functional relationship between them. The ideal form for an analog sensoris a proportional relationship:

S ¼ C þms ð38:2Þ

where C ¼ the value of the sensor output when the stimulus value is zero, and m ¼ theconstant of proportionality between s and S. The constant m indicates how much theoutput S is affected by the input s. This is referred to as the sensitivity of the measuringdevice. For example, a standard Chromel/Alumel thermocouple produces 40.6 micro-volts per �C change in temperature.

A binary sensor (e.g., limit switch, photoelectric switch) exhibits a binary relation-ship between stimulus and sensor output:

S ¼ 1 if s > 0 and S ¼ 0 if s � 0 ð38:3Þ

Before a measuring device can be used, it must be calibrated, which basically meansdetermining the transfer function of the sensor; specifically, how is the value of thestimulus s determined from the value of the output signal S? Ease of calibration is onecriterion by which a measuring device can be selected. Other criteria include accuracy,precision, operating range, speed of response, reliability and cost.

38.2.2 ACTUATORS

In automated systems, an actuator is a device that converts a control signal into a physicalaction, which usually refers to a change in a process input parameter. The action istypically mechanical, such as a change in position of a worktable or rotational speed of amotor. The control signal is generally a low level signal, and an amplifier may be requiredto increase the power of the signal to drive the actuator.

Actuators can be classified according to type of amplifier as (1) electrical, (2) hy-draulic, or (3) pneumatic. Electrical actuators include AC and DC electric motors, steppermotors, and solenoids. The operations of two types of electric motors (servomotors andsteppermotors) are described in Section 38.3.2, which deals with the analysis of positioningsystems. Hydraulic actuators utilize hydraulic fluid to amplify the control signal and areoften specified when large forces are required in the application. Pneumatic actuators aredriven by compressed air, which is commonly used in factories. All three actuator types areavailable as linear or rotational devices. This designation distinguishes whether the outputaction is a linear motion or a rotational motion. Electric motors and stepper motors aremore common as rotational actuators, whereas most hydraulic and pneumatic actuatorsprovide a linear output.

38.2.3 INTERFACE DEVICES

Interface devices allow the process to be connected to the computer controller and viceversa. Sensor signals from the manufacturing process are fed into the computer, and

Section 38.2/Hardware Components for Automation 891

E1C38 11/09/2009 18:1:2 Page 892

command signals are sent to actuators that operate the process. In this section, we discussthe hardware devices that enable this communication between the process and thecontroller. The devices include analog-to-digital converters, digital-to-analog converters,contact input/output interfaces, and pulse counters and generators.

Continuous analog signals from sensors attached to the process must be trans-formed into digital values that can be used by the control computer, a function that isaccomplished by an analog-to-digital converter (ADC). As illustrated in Figure 38.4,an ADC (1) samples the continuous signal at periodic intervals, (2) converts thesampled data into one of a finite number of defined amplitude levels, and (3) encodeseach amplitude level into a sequence of binary digits that can be interpreted by thecontrol computer. Important characteristics of an analog-to-digital converter includesampling rate and resolution. Sampling rate is the frequency with which the continuoussignal is sampled. A faster sampling rate means that the actual form of the continuoussignal can be more closely approximated. Resolution refers to the precision with whichthe analog value can be converted into binary code. This depends on the number of bitsused in the encoding procedure, the more bits, the higher the resolution. Un-fortunately, using more bits requires more time to make the conversion, which canimpose a practical limit on the sampling rate.

A digital-to-analog converter (DAC) accomplishes the reverse process of theADC. It converts the digital output of the control computer into a quasi-continuoussignal capable of driving an analog actuator or other analog device. TheDACperforms itsfunction in two steps: (1) decoding, in which the sequence of digital output values istransformed into a corresponding series of analog values at discrete time intervals, and(2) data holding, in which each analog value is changed into a continuous signal duringthe duration of the time interval. In the simplest case, the continuous signal consists of aseries of step functions, as in Figure 38.5, which are used to drive the analog actuator.

FIGURE 38.4An analog-to-digitalconverter works byconverting a continuousanalog signal into a series

of discrete sampled data.

Discretesampled signal

Continuous analog signal

Variable

Time

FIGURE 38.5An analog-to-digitalconverter works byconverting a continuous

analog signal into a seriesof discrete sampled data. Time

Series of discretestep functions

Ideal output envelopeParameter


E1C38 11/09/2009 18:1:2 Page 893

Many automated systems operate by turning on and off motors, switches, and otherdevices to respond to conditions and as a function of time. These control devices usebinary variables. They can have either of two possible values, 1 or 0, interpreted as On orOff, object present or not present, high or low voltage level, and so on. Binary sensorscommonly used in process control systems include limit switches and photocells.Common binary actuators solenoids, valves, clutches, lights, control relays, and certainmotors.

Contact input/output interfaces are components used to communicate binary databack and forth between the process and the control computer. A contact input interfaceis a device that reads binary data into the computer from an external source. It consists ofa series of binary electrical contacts that indicate the status of a binary device such as alimit switch attached to the process. The status of each contact is periodically scanned bythe computer to update values used by the control program. A contact output interface isa device used to communicate on/off signals from the computer to external binarycomponents such as solenoids, alarms, and indicator lights. It can also be used to turn onand off constant speed motors.

Asmentioned earlier, discrete data sometimes exist in the form of a series of pulses.For example, an optical encoder (discussed in Section 38.3.2) emits its measurement ofposition and velocity as a series of pulses. A pulse counter is a device that converts a seriesof pulses from an external source into a digital value, which is entered into the controlcomputer. In addition to reading the output of an optical encoder, applications of pulsecounters include counting the number of parts flowing along a conveyor past a photo-electric sensor. The opposite of a pulse counter is a pulse generator, a device thatproduces a series of electrical pulses based on digital values generated by a controlcomputer. Both the number and frequency of the pulses are controlled. An importantpulse generator application is to drive stepper motors, which respond to each step byrotating through a small incremental angle, called a step angle.

38.2.4 PROCESS CONTROLLERS

Most process control systems use some type of digital computer as the controller.Whether control involves continuous or discrete parameters and variables, or acombination of continuous and discrete, a digital computer can be connected tothe process to communicate and interact with it using the interface devices discussedin Section 38.2.3. Requirements generally associated with real-time computer controlinclude the following:

� The capability of the computer to respond to incoming signals from the process and ifnecessary, to interrupt execution of a current program to service the incoming signal.

� The capability to transmit commands to the process that are implemented by meansof actuators connected to the process. These commands may be the response toincoming signals from the process.

� The capability to execute certain actions at specific points in time during processoperation.

� The capability to communicate and interact with other computers that may beconnected to the process. The term distributed process control is used to describe acontrol system in which multiple microcomputers are used to share the processcontrol workload.

� The capability to accept input from operating personnel for purposes such as enteringnew programs or data, editing existing programs, and stopping the process in anemergency.

Section 38.2/Hardware Components for Automation 893

E1C38 11/09/2009 18:1:2 Page 894

A widely used process controller that satisfies these requirements is a program-mable logic controller. A programmable logic controller (PLC) is a microcomputer-based controller that uses stored instructions in programmable memory to implementlogic, sequencing, timing, counting, and arithmetic control functions, through digital oranalog input/output modules, for controlling various machines and processes. The majorcomponents of a PLC, shown in Figure 38.6, are (1) input and output modules, whichconnect the PLC to the industrial equipment to be controlled; (2) processor—the centralprocessing unit (CPU), which executes the logic and sequencing functions to control theprocess by operating on the input signals and determining the proper output signalsspecified by the control program; (3) PLC memory, which is connected to the processorand contains the logic and sequencing instructions; (4) power supply—115 V AC istypically used to drive the PLC. In addition, (5) a programming device (usuallydetachable) is used to enter the program into the PLC.

Programming involves entry of the control instructions to the PLC using theprogramming device. The most common control instructions include logical operations,sequencing, counting, and timing. Many control applications require additional instruc-tions for analog control, data processing, and computations. A variety of PLC program-ming languages have been developed, ranging from ladder logic diagrams to structuredtext. A discussion of these languages is beyond the scope of this text, and the reader isreferred to our references.

Advantages associated with programmable logic controllers include (1) program-ming a PLC is easier than wiring a relay control panel; (2) PLCs can be reprogrammed,whereas conventional hard-wired controls must be rewired and are often scrappedinstead because of the difficulty in rewiring; (3) a PLC can be interfaced with the plantcomputer system more readily than conventional controls; (4) PLCs require less floorspace than relay controls, and (5) PLCs offer greater reliability and easier maintenance.

38.3 COMPUTER NUMERICAL CONTROL

Numerical control (NC) is a form of programmable automation in which the mechanicalactions of a piece of equipment are controlled by a program containing coded alphanu-meric data. The data represent relative positions between a workhead and a workpart.The workhead is a tool or other processing element, and the workpart is the object beingprocessed. The operating principle of NC is to control the motion of the workhead

FIGURE 38.6 Majorcomponents of aprogrammable logic

controller.

External power source

Outputs toprocess

Inputs fromprocess

(3)

(2)

(4)

(5)

(1)Programming

device

Power supply

Memory

Inputmodule

Outputmodule

Processor


E1C38 11/09/2009 18:1:2 Page 895

relative to the workpart and to control the sequence in which the motions are carried out.The first application of numerical control was in machining (Historical Note 38.1), andthis is still an important application area. NC machine tools are shown in Figures 22.26and 22.27. Our video clip on computer numerical control shows the various types of CNCmachines and operations.

VIDEO CLIP

Computer Numerical Control. The clip contains two segments: (1) computer numericalcontrols and (2) CNC principles.

38.3.1 THE TECHNOLOGY OF NUMERICAL CONTROL

In this section we define the components of a numerical control system, and then proceedto describe the coordinate axis system and motion controls.

Components of an NC System A numerical control system consists of three basiccomponents: (1) part program, (2) machine control unit, and (3) processing equipment.The part program (the term commonly used in machine tool technology) is the detailedset of commands to be followed by the processing equipment. It is the program ofinstructions in the NC control system. Each command specifies a position or motion thatis to be accomplished by the work head relative to the workpart. A position is defined byits x-y-z coordinates. In machine tool applications, additional details in the NC programinclude spindle rotation speed, spindle direction, feed rate, tool change instructions, and

Historical Note 38.1 Numerical control [3], [5]

The initial development work on numerical control iscredited to John Parsons and Frank Stulen at the ParsonsCorporation in Michigan in the late 1940s. Parsons was amachining contractor for the U.S. Air Force and haddevised a means of using numerical coordinate data tomove the worktable of a milling machine for producingcomplex parts for aircraft. On the basis of Parson’s work,the Air Force awarded a contract to the company in 1949to study the feasibility of the new control concept formachine tools. The project was subcontracted to theMassachusetts Institute of Technology to develop aprototype machine tool that utilized the new numericaldata principle. The M.I.T. study confirmed that the conceptwas feasible and proceeded to adapt a three-axis verticalmilling machine using combined analog-digital controls.The name numerical control (NC) was given to the systemby which the machine tool motions were accomplished.The prototype machine was demonstrated in 1952.

The accuracy and repeatability of the NC systemwas far better than the manual machining methods

then available. The potential for reducingnonproductive time in the machining cycle was alsoapparent. In 1956, the Air Force sponsored thedevelopment of NC machine tools at several differentcompanies. These machines were placed in operationat various aircraft plants between 1958 and 1960. Theadvantages of NC soon became clear, and aerospacecompanies began placing orders for new NCmachines.

The importance of part programming was clear fromthe start. The Air Force continued to encourage thedevelopment and application of NC by sponsoringresearch at M.I.T. for a part programming language tocontrol NC machines. This research resulted in thedevelopment of APT in 1958 (APT stands forAutomatically Programmed Tooling). APT is a partprogramming language by which a user could write themachining instructions in simple English-like statements,and the statements were coded to be interpreted by theNC system.

Section 38.3/Computer Numerical Control 895

E1C38 11/09/2009 18:1:2 Page 896

other commands related to the operation. The part program is prepared by a partprogrammer, a person who is familiar with the details of the programming language andalso understands the technology of the processing equipment.

The machine control unit (MCU) in modern NC technology is a microcomputerthat stores and executes the program by converting each command into actions by theprocessing equipment, one command at a time. The MCU consists of both hardware andsoftware. The hardware includes the microcomputer, components to interface with theprocessing equipment, and certain feedback control elements. The software in the MCUincludes control system software, calculation algorithms, and translation software toconvert the NC part program into a usable format for the MCU. The MCU also permitsthe part program to be edited in case the program contains errors, or changes in cuttingconditions are required. Because the MCU is a computer, the term computer numericalcontrol (CNC) is often used to distinguish this type of NC from its technologicalpredecessors that were based entirely on hard-wired electronics.

The processing equipment accomplishes the sequence of processing steps totransform the starting workpart into a completed part. It operates under the controlof the MCU according to the instructions in the part program. We survey the variety ofapplications and processing equipment in Section 38.3.4.

Coordinate System and Motion Control in NC A standard coordinate axis system isused to specify positions in numerical control. The system consists of the three linear axes(x, y, z) of the Cartesian coordinate system, plus three rotational axes (a, b, c), as shown inFigure 38.7(a). The rotational axes are used to rotate the workpart to present differentsurfaces for machining, or to orient the tool or workhead at some angle relative to thepart. Most NC systems do not require all six axes. The simplest NC systems (e.g., plotters,pressworking machines for flat sheet-metal stock, and component insertion machines)are positioning systems whose locations can be defined in an x-y plane. Programming ofthese machines involves specifying a sequence of x-y coordinates. By contrast, somemachine tools have five-axis control to shape complex workpart geometries. Thesesystems typically include three linear axes plus two rotational axes.

The coordinates for a rotational NC system are illustrated in Figure 38.7(b). Thesesystems are associated with turning operations on NC lathes. Although the work rotates,this is not one of the controlled axes in a conventional NC turning system. The cuttingpath of the tool relative to the rotating workpiece is defined in the x-z plane, as shown inour figure.

FIGURE 38.7 Coordinate systems used in numerical control: (a) for flat and prismatic work, and (b) forrotational work.


E1C38 11/09/2009 18:1:2 Page 897

In many NC systems, the relative movements between the processing tool and theworkpart are accomplished by fixing the part to a worktable and then controlling thepositions and motions of the table relative to a stationary or semistationary workhead.Most machine tools and component insertion machines are based on this method ofoperation. In other systems, the workpart is held stationary and the work head is movedalong two or three axes. Flame cutters, x-y plotters, and coordinate measuring machinesoperate in this mode.

Motion control systems based on NC can be divided into two types: (1) point-to-point and (2) continuous path.Point-to-point systems, also called positioning systems,move the workhead (or workpiece) to a programmed location with no regard for thepath taken to get to that location. Once the move is completed, some processingaction is accomplished by the workhead at the location, such as drilling or punching ahole. Thus, the program consists of a series of point locations at which operations areperformed.

Continuous path systems provide continuous simultaneous control of more thanone axis, thus controlling the path followed by the tool relative to the part. This permitsthe tool to perform a process while the axes are moving, enabling the system to generateangular surfaces, two-dimensional curves, or three-dimensional contours in the workpart.This operating scheme is required in drafting machines, certain milling and turningoperations, and flame cutting. In machining, continuous path control also goes by thename contouring.

An important aspect of continuous path motion is interpolation, which isconcerned with calculating the intermediate points along a path to be followed bythe workhead relative to the part. Two common forms of interpolation are linear andcircular. Linear interpolation is used for straight line paths, in which the part pro-grammer specifies the coordinates of the beginning point and end point of the straightline as well as the feed rate to be used. The interpolator then computes the travel speedsof the two or three axes that will accomplish the specified trajectory. Circularinterpolation allows the workhead to follow a circular arc by specifying the coordinatesof its beginning and end points together with either the center or radius of the arc. Theinterpolator computes a series of small straight line segments that will approximate thearc within a defined tolerance.

Another aspect of motion control is concerned with whether the positions in thecoordinate system are defined absolutely or incrementally. In absolute positioning,the workhead locations are always defined with respect to the origin of the axis system. Inincremental positioning, the next workhead position is defined relative to the presentlocation. The difference is illustrated in Figure 38.8.

FIGURE 38.8 Absolute

vs. incremental position-ing. The workhead is atpoint (2,3) and is to be

moved to point (6,8). Inabsolute positioning, themove is specified by x¼ 6,y¼ 8;while in incremental

positioning, the move isspecified by x ¼ 4, y ¼ 5.


E1C38 11/09/2009 18:1:2 Page 898

38.3.2 ANALYSIS OF NC POSITIONING SYSTEMS

The function of the positioning system is to convert the coordinates specified in the NCpart program into relative positions between the tool and workpart during processing.Let us consider how a simple positioning system, shown in Figure 38.9, might operate.The system consists of a worktable on which a workpart is fixtured. The purpose of thetable is to move the part relative to a tool or workhead. To accomplish this purpose, theworktable is moved linearly by means of a rotating leadscrew that is driven by a motor.For simplicity, only one axis is shown in our sketch. To provide x-y capability, thesystem shown would be piggybacked on top of a second axis perpendicular to the first.The leadscrew has a certain pitch p, mm/thread (in/thread) or mm/rev (in/rev). Thus,the table is moved a distance equal to the leadscrew pitch for each revolution. Thevelocity at which the worktable moves is determined by the rotational speed of theleadscrew.

Two basic types of motion control are used in NC: (a) open loop and (b) closedloop, as shown in Figure 38.10. The difference is that an open-loop system operates

FIGURE 38.9 Motorandleadscrewarrangement inan NC positioning system.

FIGURE 38.10 Twotypes of motion control innumerical control: (a)

open loop and (b) closedloop.


E1C38 11/09/2009 18:1:2 Page 899

without verifying that the desired position of the worktable has been achieved. A closed-loop control system uses feedback measurement to verify that the position of theworktable is indeed the location specified in the program. Open-loop systems are lessexpensive than closed-loop systems and are appropriate when the force resisting theactuating motion is minimal, as in point-to-point drilling, for example. Closed-loopsystems are normally specified for machine tools that perform continuous path opera-tions such as milling or turning, in which the resisting forces can be significant.

Open-Loop Positioning Systems To turn the leadscrew, an open-loop positioningsystem typically uses a stepping motor (a.k.a. stepper motor). In NC, the stepping motoris driven by a series of electrical pulses generated by themachine control unit. Each pulsecauses the motor to rotate a fraction of one revolution, called the step angle. Theallowable step angles must conform to the relationship

a ¼ 360

nsð38:1Þ

where a ¼ step angle, degrees; and ns ¼ the number of step angles for the motor, whichmust be an integer. The angle through which the motor shaft rotates is given by

Am ¼ anp ð38:2Þwhere Am ¼ angle of motor shaft rotation, degrees; np ¼ number of pulses received by themotor; and a¼ step angle, here defined as degrees/pulse. Finally, the rotational speed of themotor shaft is determined by the frequency of pulses sent to the motor:

Nm ¼ 60af p360

ð38:3Þ

where Nm ¼ speed of motor shaft rotation, rev/min; fp ¼ frequency of pulses drivingthe stepper motor, Hz (pulses/sec), the constant 60 converts pulses/sec to pulses/min;the constant 360 converts degrees of rotation to full revolutions; and a ¼ step angleof the motor, as before.

Themotor shaft drives the leadscrew that determines the position and velocity of theworktable.Theconnection is oftendesignedusing agear reduction to increase theprecisionof table movement. However, the angle of rotation and rotational speed of the leadscreware reduced by this gear ratio. The relationships are as follows:

Am ¼ rgAls ð38:4aÞand

Nm ¼ rgNls ð38:4bÞ

whereAm andNm are the angle of rotation, degrees, and rotational speed, rev/min, of themotor, respectively; Als and Nls are the angle of rotation, degrees, and rotational speed,rev/min, of the leadscrew, respectively; and rg ¼ gear reduction between the motor shaftand the leadscrew; for example, a gear reduction of 2 means that the motor shaft rotatesthrough two revolutions for each rotation of the leadscrew.

The linear position of the table in response to the rotation of the leadscrew dependson the leadscrew pitch p, and can be determined as follows:

x ¼ pAls

360ð38:5Þ

where x ¼ x-axis position relative to the starting position, mm (in); p ¼ pitch of theleadscrew, mm/rev (in/rev); and Als/360 ¼ the number of revolutions (and partialrevolutions) of the leadscrew. By combining Eqs. (38.2), (38.4a), and (38.5) and


E1C38 11/09/2009 18:1:2 Page 900

rearranging, the number of pulses required to achieve a specified x-position increment ina point-to-point system can be found:

np ¼ 360rgx

pa¼ rgnsAls

360ð38:6Þ

The velocity of the worktable in the direction of the leadscrew axis can be determined asfollows:

vt ¼ f r ¼ Nlsp ð38:7Þ

where vt ¼ table travel speed, mm/min (in/min); fr ¼ table feed rate, mm/min (in/min);Nls ¼ rotational speed of the leadscrew, rev/min; and p ¼ leadscrew pitch, mm/rev (in/rev). The rotational speed of the leadscrew depends on the frequency of pulses drivingthe stepping motor:

Nls ¼60f pnsrg

ð38:8Þ

where Nls ¼ leadscrew rotational speed, rev/min; fp ¼ pulse train frequency, Hz (pulses/sec); ns ¼ steps/rev, or pulses/rev, and rg ¼ gear reduction between the motor and theleadscrew. For a two-axis table with continuous path control, the relative velocities of theaxes are coordinated to achieve the desired travel direction. Finally, the required pulsefrequency to drive the table at a specified feed rate can be obtained by combining Eqs.(38.7) and (38.8) and rearranging to solve for fp:

f p ¼ vtnsrg60p

¼ f rnsrg60p

¼ Nlsnsrg60

¼ Nmns60

ð38:9Þ

Example 38.1Open-LoopPositioning

A stepping motor has 48 step angles. Its output shaft is coupled to a leadscrew with a 4:1gear reduction (four turns of the motor shaft for each turn of the leadscrew). Theleadscrew pitch ¼ 5.0 mm. The worktable of a positioning system is driven by theleadscrew. The table must move a distance of 75.0 mm from its current position at a travelspeed of 400 mm/min. Determine (a) howmany pulses are required to move the table thespecified distance and (b) the motor speed and (c) pulse frequency required to achievethe desired table speed.

Solution: (a) Tomove a distance x¼ 75mm, the leadscrewmust rotate through an anglecalculated as follows:

Als ¼ 360x

p¼ 360 75ð Þ

5¼ 5400�

With 48 step angles and a gear reduction of 4, the number of pulses to move the table 75mm is

np ¼ 4 48ð Þ 5400ð Þ360

¼ 2880 pulses

(b) Equation (38.7) can be used to find the leadscrew speed corresponding to the tablespeed of 400 mm/min,

Nls ¼ vtp¼ 400

5:0¼ 80:0 rev/min


E1C38 11/09/2009 18:1:3 Page 901

The motor speed will be 4 times as fast:

Nm ¼ rgNls ¼ 4 80ð Þ ¼ 320 rev/min

(c) Finally, the pulse rate is given by Eq. (38.13):

f p ¼ 320 48ð Þ60

¼ 256 Hzn

Closed-Loop Positioning Systems Closed-loop NC systems, Figure 38.10(b), useservomotors and feedback measurements to ensure that the desired position is achieved.A common feedback sensor used in NC (and also industrial robots) is the optical rotaryencoder, illustrated in Figure 38.11. It consists of a light source, a photocell, and a diskcontaining a series of slots through which the light source can shine to energize thephotocell. The disk is connected to a rotating shaft, which in turn is connected directly tothe leadscrew. As the leadscrew rotates, the slots cause the light source to be seen by thephotocell as a series of flashes, which are converted into an equivalent series of electricalpulses. By counting the pulses and computing the frequency of the pulse train, theleadscrew angle and rotational speed can be determined, and thus worktable position andspeed can be calculated using the pitch of the leadscrew.

The equations describing the operation of a closed-loop positioning system areanalogous to those for an open-loop system. In the basic optical encoder, the anglebetween slots in the disk must satisfy the following requirement:

a ¼ 360

nsð38:10Þ

where a¼ angle between slots, degrees/slot; and ns¼ the number of slots in the disk, slots/rev; and 360 ¼ degrees/rev. For a certain angular rotation of the leadscrew, the encodergenerates a number of pulses given by

np ¼ Als

a¼ Alsns

360ð38:11Þ

where np ¼ pulse count; Als ¼ angle of rotation of the leadscrew, degrees; and a ¼ anglebetween slots in the encoder, degrees/pulse. The pulse count can be used to determine thelinear x-axis position of the worktable by factoring in the leadscrew pitch. Thus,

x ¼ pnpns

¼ pAls

360ð38:12Þ

FIGURE 38.11 Optical

encoder: (a) apparatus,and (b) series of pulsesemitted to measurerotation of disk.


E1C38 11/09/2009 18:1:3 Page 902

Similarly, the feed rate at which the worktable moves is obtained from the frequency ofthe pulse train:

vt ¼ f r ¼60 pf pns

ð38:13Þ

where vt ¼ table travel speed, mm/min (in/min); fr ¼ feed rate, mm/min (in/min);p ¼ pitch, mm/rev (in/rev); fp ¼ frequency of the pulse train, Hz (pulses/sec); ns ¼number of slots in the encoder disk, pulses/rev; and 60 converts seconds to minutes.The speed relationship given by Eq. (38.7) is also valid for a closed-loop positioningsystem.

The series of pulses generated by the encoder is compared with the coordinateposition and feed rate specified in the part program, and the difference is used by themachine control unit to drive a servomotor that in turn drives the leadscrew andworktable.As with the open-loop system, a gear reduction between the servomotor and the leadscrewcan also be used, so Eqs. (38.4) are applicable. A digital-to-analog converter is used toconvert the digital signals used by the MCU into a continuous analog signal to operate thedrive motor. Closed-loop NC systems of the type described here are appropriate whenthere is force resisting the movement of the table. Most metal-machining operations fallinto this category, particularly those involving continuous path control such as milling andturning.

Example 38.2 NCClosed-LoopPositioning

An NC worktable is driven by a closed-loop positioning system consisting of aservomotor, leadscrew, and optical encoder. The leadscrew has a pitch ¼ 5.0 mmand is coupled to the motor shaft with a gear ratio of 4:1 (four turns of the motor foreach turn of the leadscrew). The optical encoder generates 100 pulses/rev of theleadscrew. The table has been programmed to move a distance of 75.0 mm at a feedrate¼ 400 mm/min. Determine (a) howmany pulses are received by the control systemto verify that the table has moved exactly 75.0 mm; and (b) the pulse rate and (c) motorspeed that correspond to the specified feed rate.

Solution: (a) Rearranging Eq. (38.12) to find np,

np ¼ xnsp

¼ 75 100ð Þ5

¼ 1500 pulses

(b) The pulse rate corresponding to 400 mm/min can be obtained by rearrangingEq. (38.13):

f p ¼ f rns60p

¼ 400 100ð Þ60 5ð Þ ¼ 133:33Hz

(c) Leadscrew rotational speed is the table velocity divided by the pitch:

Nls ¼ f rp¼ 80 rev/min

With a gear ratio rg ¼ 4.0, the motor speed N ¼ 4 80ð Þ ¼ 320 rev/min n

Precision in Positioning Three critical measures of precision in positioning are controlresolution, accuracy, and repeatability. These terms are most easily explained by con-sidering a single axis of the position system.


E1C38 11/09/2009 18:1:3 Page 903

Control resolution refers to the system’s ability to divide the total range of the axismovement into closely spaced points that can be distinguished by the control unit.Control resolution is defined as the distance separating two adjacent control points inthe axis movement. Control points are sometimes called addressable points becausethey are locations along the axis to which the worktable can be directed to go. It isdesirable for the control resolution to be as small as possible. This depends onlimitations imposed by (1) the electromechanical components of the positioningsystem, and/or (2) the number of bits used by the controller to define the axis coordinatelocation.

The electromechanical factors that limit resolution include leadscrew pitch, gearratio in thedrive system,and thestepangle ina steppingmotor (for anopen-loopsystem)orthe angle between slots in an encoder disk (for a closed-loop system). Together, thesefactors determine a control resolution, or minimum distance that the worktable can bemoved. For example, the control resolution for an open-loop system driven by a steppermotor with a gear reduction between the motor shaft and the leadscrew is given by

CR1 ¼ p

nsrgð38:14aÞ

where CR1 ¼ control resolution of the electromechanical components, mm (in); p ¼leadscrew pitch, mm/rev (in/rev); ns ¼ number of steps/rev; and rg ¼ gear reduction.

The corresponding expression for a closed-loop positioning system is similar butdoes not include the gear reduction because the encoder is connected directly to theleadscrew. There is no gear reduction. Thus, control resolution for a closed-loop system isdefined as follows:

CR1 ¼ p

nsð38:14bÞ

where ns in this case refers to the number of slots in the optical encoder.Although unusual in modern computer technology, the second possible factor that

could limit control resolution is the number of bits defining the axis coordinate value. Forexample, this limitation may be imposed by the bit storage capacity of the controller. IfB ¼ the number of bits in the storage register for the axis, then the number of controlpoints into which the axis range can be divided¼ 2B. Assuming that the control points areseparated equally within the range, then

CR2 ¼ L

2B � 1ð38:15Þ

where CR2 ¼ control resolution of the computer control system, mm (in); and L ¼ axisrange, mm (in). The control resolution of the positioning system is the maximum of thetwo values; that is,

CR ¼ Max CR1;CR2f g ð38:16ÞIt is generally desirable forCR2�CR1, meaning that the electromechanical system is thelimiting factor in control resolution.

When a positioning system is directed to move the worktable to a given controlpoint, the capability of the system to move to that point will be limited by mechanicalerrors. These errors are due to a variety of inaccuracies and imperfections in themechanical system, such as play between the leadscrew and the worktable, backlashin the gears, and deflection of machine components. It is convenient to assume that theerrors form a statistical distribution about the control point that is an unbiased normaldistribution with mean ¼ 0. If we further assume that the standard deviation of thedistribution is constant over the range of the axis under consideration, then nearly all ofthe mechanical errors (99.73%) are contained within �3 standard deviations of the


E1C38 11/09/2009 18:1:3 Page 904

control point. This is pictured in Figure 38.12 for a portion of the axis range, whichincludes three control points.

Given these definitions of control resolution and mechanical error distribution, letus now consider accuracy and repeatability. Accuracy is defined in a worst-case scenarioin which the desired target point lies exactly between two adjacent control points. Sincethe system can only move to one or the other of the control points, there will be an errorin the final position of the worktable. If the target were closer to one of the control points,then the table would be moved to the closer control point and the error would be smaller.It is appropriate to define accuracy in the worst case. The accuracy of any given axis of apositioning system is the maximum possible error that can occur between the desiredtarget point and the actual position taken by the system; in equation form,

Accuracy ¼ 0:5CRþ 3s ð38:17Þ

where CR ¼ control resolution, mm (in); and s ¼ standard deviation of the errordistribution, mm (in).

Repeatability refers to the capability of a positioning system to return to a givencontrol point that has been previously programmed. This capability can be measured interms of the location errors encountered when the system attempts to position itself at thecontrol point. Location errors are a manifestation of the mechanical errors of thepositioning system, which are defined by an assumed normal distribution, as describedabove. Thus, the repeatability of any given axis of a positioning system can be defined asthe range of mechanical errors associated with the axis; this reduces to

Repeatability ¼ �3s ð38:18Þ

Example 38.3ControlResolution,Accuracy, andRepeatability

Referring back to Example 38.1, themechanical inaccuracies in the open-loop positioningsystem can be described by a normal distribution whose standard deviation ¼ 0.005 mm.The range of the worktable axis is 550mm, and there are 16 bits in the binary register usedby the digital controller to store the programmed position. Determine (a) controlresolution, (b) accuracy, and (c) repeatability for the positioning system.

Solution: (a) Control resolution is the greater of CR1 and CR2 as defined by Eqs.(38.14a) and (38.15):

CR1 ¼ p

nsrg¼ 5:0

48 4ð Þ ¼ 0:0260mm

CR2 ¼ L

2B � 1¼ 550

216 � 1¼ 550

65; 535¼ 0:0084mm

CR ¼ Max 0:0260; 0:0084f g ¼ 0:0260mm

FIGURE 38.12A portion of a linearpositioning system axis,

with definition of controlresolution, accuracy, andrepeatability.

Control resolution= CR

Repeatablity = ±3

Axis

CR + 312

Accuracy =

Controlpoint

Controlpoint

Desired targetpoint

Distribution ofmechanical errors


E1C38 11/09/2009 18:1:3 Page 905

(b) Accuracy is given by Eq. (38.17):

Accuracy ¼ 0:5 0:0260ð Þ þ 3 0:005ð Þ ¼ 0:0280mm

(c) Repeatability ¼ � 3(0.005) ¼ � 0.015 mm. n

38.3.3 NC PART PROGRAMMING

In machine tool applications, the task of programming the system is called NC partprogramming because the program is prepared for a given part. It is usually accom-plished by someone familiar with the metalworking process who has learned theprogramming procedure for the particular equipment in the plant. For other processes,other terms may be used for programming, but the principles are similar and a trainedindividual is needed to prepare the program. Computer systems are used extensively toprepare NC programs.

Part programming requires the programmer to define the points, lines, and surfaces ofthe workpart in the axis system, and to control the movement of the cutting tool relative tothese defined part features. Several part programming techniques are available, the mostimportant of which are (1) manual part programming, (2) computer-assisted part program-ming, (3) CAD/CAM-assisted part programming, and (4) manual data input.

Manual Part Programming For simple point-to-point machining jobs, such as drillingoperations, manual programming is often the easiest and most economical method.Manual part programming uses basic numerical data and special alphanumeric codes todefine the steps in the process. For example, to perform a drilling operation, a commandof the following type is entered:

n010x70:0 y85:5 f175 s500

Each ‘‘word’’ in the statement specifies a detail in the drilling operation.Then-word (n010)is simply a sequence number for the statement. The x- and y-words indicate the x and ycoordinate positions (x ¼ 70.0 mm and y ¼ 85.5 mm). The f-word and s-word specify thefeed rate and spindle speed tobeused in thedrilling operation (feed rate¼ 175mm/minandspindle speed ¼ 500 rev/min). The complete NC part program consists of a sequence ofstatements similar to the above command.

Computer-Assisted Part Programming Computer-assisted part programming in-volves the use of a high-level programming language. It is suited to the programmingof more complex jobs than manual programming. The first part programming languagewas APT (Automatically Programmed Tooling), developed as an extension of theoriginal NC machine tool research and first used in production around 1960.

In APT, the part programming task is divided into two steps: (1) definition of partgeometry and (2) specification of tool path and operation sequence. In step 1, the partprogrammer defines the geometry of the workpart by means of basic geometric elementssuch as points, lines, planes, circles, and cylinders. These elements are defined using APTgeometry statements, such as

P1 ¼ POINT=25:0; 150:0

L1 ¼ LINE=P1; P2

P1 is a point defined in the x-y plane located at x ¼ 25 mm and y ¼ 150 mm. L1 is a linethat goes through points P1 and P2. Similar statements can be used to define circles,


E1C38 11/09/2009 18:1:3 Page 906

cylinders, and other geometry elements. Most workpart shapes can be described usingstatements like these to define their surfaces, corners, edges, and hole locations.

Specification of the tool path is accomplished with APT motion statements.A typical statement for point-to-point operation is

GOTO=P1

This directs the tool to move from its current location to a position defined by P1,where P1 has been defined by a previous APT geometry statement. Continuous pathcommands use geometry elements such as lines, circles, and planes. For example, thecommand

GORGT=L3; PAST; L4

directs the tool to go right (GORGT) along line L3 until it is positioned just past line L4(of course, L4 must be a line that intersects L3).

Additional APT statements are used to define operating parameters such as feedrates, spindle speeds, tool sizes, and tolerances. When completed, the part programmerenters the APT program into the computer, where it is processed to generate low-levelstatements (similar to statements prepared in manual part programming) that can beused by a particular machine tool.

CAD/CAM-Assisted Part Programming The use of CAD/CAM takes computer-assisted part programming a step further by using a computer graphics system (CAD/CAM system) to interact with the programmer as the part program is being prepared. Inthe conventional use of APT, a complete program is written and then entered into thecomputer for processing. Many programming errors are not detected until computerprocessing.When aCAD/CAMsystem is used, the programmer receives immediate visualverification when each statement is entered, to determine whether the statement iscorrect. When part geometry is entered by the programmer, the element is graphicallydisplayed on the monitor. When the tool path is constructed, the programmer can seeexactly how the motion commands will move the tool relative to the part. Errors can becorrected immediately rather than after the entire program has been written.

Interaction between programmer and programming system is a significant benefitof CAD/CAM-assisted programming. There are other important benefits of using CAD/CAM in NC part programming. First, the design of the product and its components mayhave been accomplished on a CAD/CAM system. The resulting design database,including the geometric definition of each part, can be retrieved by the NC programmerto use as the starting geometry for part programming. This retrieval saves valuable timecompared to reconstructing the part from scratch using the APT geometry statements.

Second, special software routines are available in CAD/CAM-assisted part pro-gramming to automate portions of the tool path generation, such as profile millingaround the outside periphery of a part, milling a pocket into the surface of a part, surfacecontouring, and certain point-to-point operations. These routines are called by the partprogrammer as special macro commands. Their use results in significant savings inprogramming time and effort.

Manual Data Input Manual data input (MDI) is a method in which a machine operatorenters the part program in the factory. The method involves use of a CRT display withgraphics capability at the machine tool controls. NC part programming statements areentered using a menu-driven procedure that requires minimum training of the machinetool operator. Because part programming is simplified and does not require a special staffof NC part programmers, MDI is a way for small machine shops to economicallyimplement numerical control into their operations.


E1C38 11/09/2009 18:1:3 Page 907

38.3.4 APPLICATIONS OF NUMERICAL CONTROL

Machining is an important application area for numerical control, but the operatingprinciple of NC can be applied to other operations as well. There are many industrialprocesses in which the position of a workhead must be controlled relative to the part orproduct being worked on. We divide the applications into two categories: (1) machinetool applications, and (2) nonmachine tool applications. It should be noted that theapplications are not all identified by the name numerical control in their respectiveindustries.

In the machine tool category, NC is widely used for machining operations such asturning, drilling, and milling (Sections 22.2, 22.3, and 22.4, respectively). The use of NC inthese processes has motivated the development of highly automatedmachine tools calledmachining centers, which change their own cutting tools to perform a variety ofmachining operations under NC program control (Section 22.5). In addition to machin-ing, other numerically controlled machine tools include (1) grinding machines (Section25.1); (2) sheet metal pressworkingmachines (Section 20.5.2); (3) tube-bendingmachines(Section 20.7); and (4) thermal cutting processes (Section 26.3).

In the nonmachine tool category, NC applications include (1) tape-laying machinesand filament-winding machines for composites (Section 15.2.3 and Section 15.4);(2) welding machines, both arc welding (Section 31.1) and resistance welding (Section31.2); (3) component-insertionmachines in electronics assembly (Sections 35.3 and 35.4);(4) drafting machines; and (5) coordinate measuring machines for inspection (Section42.6.1).

Benefits of NC relative to manually operated equipment in these applicationsinclude (1) reduced nonproductive time, which results in shorter cycle times, (2) lowermanufacturing lead times, (3) simpler fixturing, (4) greater manufacturing flexibility, (5)improved accuracy, and (6) reduced human error.

38.4 INDUSTRIAL ROBOTICS

An industrial robot is a general-purpose programmable machine possessing certain anthro-pomorphic features. The most obvious anthropomorphic, or human-like, feature is therobot’s mechanical arm, or manipulator. The control unit for a modern industrial robot is acomputer that can be programmed to execute rather sophisticated subroutines, thusproviding the robot with an intelligence that sometimes seems almost human. The robot’smanipulator, combined with a high-level controller, allows an industrial robot to perform avariety of tasks such as loading and unloading production machine, spot welding, and spraypainting. Robots are typically used as substitutes for human workers in these tasks. The firstindustrial robotwas installed in a die-casting operation at FordMotorCompany. The robot’sjob was to unload die castings from the die-casting machine.

In this section, we consider various aspects of robot technology and applications,including how industrial robots are programmed to perform their tasks.

38.4.1 ROBOT ANATOMY

An industrial robot consists of a mechanical manipulator and a controller to move it andperform other related functions. The mechanical manipulator consists of joints and linksthat can position and orient the end of the manipulator relative to its base. The controllerunit consists of electronic hardware and software to operate the joints in a coordinatedfashion to execute the programmed work cycle. Robot anatomy is concerned with the

Section 38.4/Industrial Robotics 907

E1C38 11/09/2009 18:1:3 Page 908

mechanical manipulator and its construction. Figure 38.13 shows one of the commonindustrial robot configurations.

Manipulator Joints and Links A joint in a robot is similar to a joint in a human body. Itprovides relative movement between two parts of the body. Connected to each joint arean input link and an output link. Each joint moves its output link relative to its input link.The robot manipulator consists of a series of link–joint–link combinations. The outputlink of one joint is the input link for the next joint. Typical industrial robots have five orsix joints. The coordinated movement of these joints gives the robot its ability to move,position, and orient objects to perform useful work. Manipulator joints can be classifiedas linear or rotating, indicating the motion of the output link relative to the input link.

Manipulator Design Using joints of the two basic types, each joint separated from theprevious by a link, the manipulator is constructed. Most industrial robots are mounted tothe floor. We can identify the base as link 0; this is the input link to joint 1 whose output islink 1, which is the input to joint 2 whose output link is link 2; and so forth, for the numberof joints in the manipulator.

Robot manipulators can usually be divided into two sections: arm-and-bodyassembly and wrist assembly. There are typically three joints associated with the arm-and-body assembly, and two or three joints associated with the wrist. The function of the

FIGURE 38.13 The

manipulator of a modernindustrial robot. (Photocourtesy of Adept

Technology, Inc.,Pleasanton, California.)


E1C38 11/09/2009 18:1:3 Page 909

arm-and-body is to position an object or tool, and the wrist function is to properly orientthe object or tool. Positioning is concerned withmoving the part or tool from one locationto another. Orientation is concerned with precisely aligning the object relative to somestationary location in the work area.

To accomplish these functions, arm-and-body designs differ from those of the wrist.Positioning requires large spatial movements, while orientation requires twisting androtating motions to align the part or tool relative to a fixed position in the workplace. Thearm-and-body consists of large links and joints, whereas the wrist consists of short links.The arm-and-body joints often consist of both linear and rotating types, while the wristjoints are almost always rotating types.

There are five basic arm-and-body configurations available in commercial robots,identified in Figure 38.14. The design shown in part (e) of the figure and in Figure 38.13 iscalled a SCARA robot, which stands for ‘‘selectively compliant assembly robot arm.’’ Itis similar to a jointed arm anatomy, except that the shoulder and elbow joints havevertical axes of rotation, thus providing rigidity in the vertical direction but relativecompliance in the horizontal direction.

FIGURE 38.14 Five common anatomies of commercial industrial robots: (a) polar, (b) cylindrical, (c) Cartesian

coordinate, (d) jointed-arm, and (e) SCARA, or selectively compliant assembly robot arm.


E1C38 11/09/2009 18:1:4 Page 910

The wrist is assembled to the last link in any of these arm-and-body con-figurations. The SCARA is sometimes an exception because it is almost alwaysused for simple handling and assembly tasks involving vertical motions. Therefore,a wrist is not usually present at the end of its manipulator. Substituting for the wrist onthe SCARA is usually a gripper to grasp components for movement and/or assembly.

Work Volume and Precision ofMotion One of the important technical considerationsof an industrial robot is the size of its work volume. Work volume is defined as theenvelope within which a robot manipulator can position and orient the end of its wrist.This envelope is determined by the number of joints, as well as their types and ranges, andthe sizes of the links. Work volume is important because it plays a significant role indetermining which applications a robot can perform.

The definitions of control resolution, accuracy, and repeatability developed inSection 38.3.2 for NC positioning systems apply to industrial robots. A robot manipulatoris, after all, a positioning system. In general, the links and joints of robots are not nearly asrigid as their machine tool counterparts, and so the accuracy and repeatability of theirmovements are not as good.

End Effectors An industrial robot is a general-purposemachine. For a robot to be usefulin a particular application, it must be equipped with special tooling designed for theapplication. An end effector is the special tooling that connects to the robot’s wrist-end toperform the specific task. There are two general types of end effector: tools and grippers.A tool is used when the robot must perform a processing operation. The special toolsinclude spot-welding guns, arc-welding tools, spray-painting nozzles, rotating spindles,heating torches, and assembly tools (e.g., automatic screwdriver). The robot is pro-grammed to manipulate the tool relative to the workpart being processed.

Grippers are designed to grasp andmove objects during the work cycle. The objectsare usually workparts, and the end effector must be designed specifically for the part.Grippers are used for part placement applications, machine loading and unloading, andpalletizing. Figure 38.15 shows a typical gripper configuration.

38.4.2 CONTROL SYSTEMS AND ROBOT PROGRAMMING

The robot’s controller consists of the electronic hardware and software to control the jointsduring execution of a programmed work cycle. Most robot control units today are based ona microcomputer system. The control systems in robotics can be classified as follows:

FIGURE 38.15 A robotgripper: (a) open and(b) closed to grasp a

workpart.


E1C38 11/09/2009 18:1:4 Page 911

1. Playback with point-to-point (PTP) control. As in numerical control, robot motionsystems can be divided into point-to-point and continuous path. The program for apoint-to-point playback robot consists of a series of point locations and the sequence inwhich these points must be visited during the work cycle. During programming, thesepoints are recorded into memory, and then subsequently played back during executionof the program. In a point-to-point motion, the path taken to get to the final position isnot controlled.

2. Playback with continuous path (CP) control. Continuous path control is similar toPTP, exceptmotion paths rather than individual points are stored inmemory. In certaintypes of regular CP motions, such as a straight line path between two point locations,the trajectory required by the manipulator is computed by the controller unit for eachmove. For irregular continuous motions, such as a path followed in spray painting, thepath is defined by a series of closely spaced points that approximate the irregularsmooth path. Robots capable of continuous path motions can also execute point-to-point movements.

3. Intelligent control. Modern industrial robots exhibit characteristics that often makethem appear to be acting intelligently. These characteristics include the ability torespond to sophisticated sensors such asmachine vision,make decisions when things gowrong during the work cycle, make computations, and communicate with humans.Robot intelligence is implemented using powerful microprocessors and advancedprogramming techniques.

Robots execute a stored program of instructions that define the sequenceof motions and positions in the work cycle, much like a part program in NC. In additionto motion instructions, the program may include instructions for other functions such asinteracting with external equipment, responding to sensors, and processing data.

There are two basic methods used to teach modern robots their programs: lead-through programming and computer programming languages.Leadthrough programminginvolves a ‘‘teach-by-showing’’method inwhich themanipulator ismovedby the program-mer through the sequence of positions in the work cycle. The controller records eachposition inmemory for subsequentplayback.Twoprocedures for leading the robot throughthe motion sequence are available: powered leadthrough and manual leadthrough. Inpowered leadthrough, themanipulator is driven by a control box that has toggle switches orpress buttons to control the movements of the joints. Using the control box, the program-mer moves the manipulator to each location, recording the corresponding joint positionsinto memory. Powered leadthrough is the common method for programming playbackrobots with point-to-point control. Manual leadthrough is typically used for playbackrobots with continuous path control. In this method, the programmer physicallymoves themanipulatorwrist through themotion cycle. For spraypainting and certainother jobs, this isa more convenient means of programming the robot.

Computer programming languages for programming robots have evolved from theuse of microcomputer controllers. The first commercial language was introduced around1979. Computer languages provide a convenient way to integrate certain nonmotionfunctions into the work cycle, such as decision logic, interlocking with other equipment,and interfacing with sensors. A more thorough discussion of robot programming ispresented in reference [6].

38.4.3 APPLICATIONS OF INDUSTRIAL ROBOTS

Some industrial work lends itself to robot applications. The following are the importantcharacteristics of a work situation that tend to promote the substitution of a robot in placeof a humanworker: (1) the work environment is hazardous for humans, (2) the work cycle


E1C38 11/09/2009 18:1:4 Page 912

is repetitive, (3) the work is performed at a stationary location, (4) part or tool handlingwould be difficult for humans, (5) it is a multishift operation, (6) there are longproduction runs and infrequent changeovers, and (7) part positioning and orientationare established at the beginning of the work cycle, since most robots cannot see.

Applications of industrial robots that tend to match these characteristics can bedivided into three basic categories: (1) material handling, (2) processing operations, and(3) assembly and inspection.

Material handling applications involve the movement of materials or parts fromone location and orientation to another. To accomplish this relocation task, the robot isequipped with a gripper. As noted earlier, the gripper must be custom-designed tograsp the particular part in the application. Material handling applications includematerial transfer (part placement, palletizing, depalletizing) and machine loading and/or unloading (e.g., machine tools, presses, and plastic molding).

Processing operations require the robot tomanipulate a tool as its end effector. Theapplications include spot welding, continuous arc welding, spray coating, and certainmetal cutting and deburring operations in which the robot manipulates a special tool. Ineach of these operations, the tool is used as the robot’s end effector. An application ofspot welding is illustrated in Figure 38.16. Spot welding is a common application ofindustrial robots in the automotive industry.

Assembly and inspection applications cannot be classified neatly in either of theprevious categories; they sometimes involve part handling and other times manipulationof a tool. Assembly applications often involve the stacking of one part onto anotherpart—basically a part handling task. In other assembly operations a tool is manipulated,such as an automatic screwdriver. Similarly, inspection operations sometimes require therobot to position a workpart relative to an inspection device, or to load a part into aninspection machine; other applications involve the manipulation of a sensor to performan inspection.

FIGURE 38.16Aportionofanautomobileassembly line in which

robots perform spot-welding operations.(Photo courtesy of Ford

Motor Company,Dearborn, Michigan.)


E1C38 11/09/2009 18:1:5 Page 913

REFERENCES

[1] Asfahl, C. R. Robots and Manufacturing Automa-tion. John Wiley & Sons, Inc., New York, 1992.

[2] Bollinger, J. G., and Duffie N. A. Computer Controlof Machines and Processes. Addison-Wesley Long-man, Inc., New York, 1989.

[3] Chang, C-H., and Melkanoff, M. A. NC MachineProgramming and Software Design, 3rd ed. PrenticeHall, Inc., Upper Saddle River, New Jersey, 2005.

[4] Engelberger, J. F. Robotics in Practice: Manage-ment and Applications of Robotics in Industry.AMACOM, New York, 1985.

[5] Groover, M. P. Automation, Production Systems,and Computer Integrated Manufacturing, 3rd ed.Pearson/Prentice Hall, Upper Saddle River, NewJersey, 2008.

[6] Groover, M. P., Weiss, M., Nagel, R. N., and Odrey,N. G. Industrial Robotics: Technology, Program-

ming, and Applications. McGraw-Hill, New York,1986.

[7] Hughes, T. A., Programmable Controllers, 4th ed.Instrumentation, Systems, and Automation Society,Research Triangle Park, North Carolina, 2005.

[8] Pessen, D. W. Industrial Automation. John Wiley &Sons, Inc., New York, 1989.

[9] Seames W. Computer Numerical Control, Conceptsand Programming. Delmar-Thomson Learning,Albany, New York, 2002.

[10] Webb, J. W., and Reis, R. A. Programmable LogicControllers: Principles and Applications, 5th ed.Pearson/Prentice Hall, Upper Saddle River, NewJersey, 2003.

[11] Weber, A.‘‘Robot Dos and Don’ts,’’ Assembly, Feb-ruary 2005, pp. 50–57.

REVIEW QUESTIONS

38.1. Define the term manufacturing system.38.2. What are the three basic components of an auto-

mated system?38.3. What are some of the advantages of using electrical

power in an automated system?38.4. What is the difference between a closed-loop con-

trol system and an open-loop control system?38.5. What is the difference between fixed automation

and programmable automation?38.6. What is a sensor?38.7. What is an actuator in an automated system?38.8. What is a contact input interface?38.9. What is a programmable logic controller?

38.10. Identify and briefly describe the three basic com-ponents of a numerical control (NC) system.

38.11. What is the difference between point-to-point andcontinuous path in a motion control system?

38.12. What is the difference between absolute position-ing and incremental positioning?

38.13. What is the difference between an open-loop posi-tioning system and a closed-loop positioningsystem?

38.14. Under what circumstances is a closed-loop posi-tioning system preferable to an open-loop system?

38.15. Explain the operation of an optical encoder.38.16. Why should the electromechanical system rather

than the controller storage register be the limitingfactor in control resolution?

38.17. What ismanualdata input inNCpartprogramming?38.18. Identify some of the non-machine tool applications

of numerical control.38.19. What are some of the benefits usually cited for NC

compared to using manual alternative methods?38.20. What is an industrial robot?38.21. How is an industrial robot similar to numerical

control?38.22. What is an end effector?38.23. In robot programming, what is the difference between

powered leadthrough and manual leadthrough?

MULTIPLE CHOICE QUIZ

There are 21 correct answers in the following multiple choice questions (some questions have multiple answers that arecorrect). To attain a perfect score on the quiz, all correct answers must be given. Each correct answer is worth 1 point.Each omitted answer or wrong answer reduces the score by 1 point, and each additional answer beyond the correctnumber of answers reduces the score by 1 point. Percentage score on the quiz is based on the total number of correctanswers.

Multiple Choice Quiz 913

E1C38 11/09/2009 18:1:5 Page 914

38.1. The three components of an automated system arewhich of the following: (a) actuators, (b) commu-nication system, (c) control system, (d) feedbackloop, (e) humans, (f) power, (g) program of instruc-tions, and (h) sensors?

38.2. The three basic types of automated systems used inmanufacturing are fixed automation, programma-ble automation, and flexible automation. Flexibleautomation is an extension of programmable auto-mation in which there is virtually no lost produc-tion time for setup changes or reprogramming:(a) true or (b) false?

38.3. The input/output relationship of a sensor is calledwhich one of the following: (a) analog, (b) con-verter, (c) sensitivity, or (d) transfer function?

38.4. A stepper motor is which one of the following typesof devices: (a) actuator, (b) interface device,(c) pulse counter, or (d) sensor?

38.5. A contact input interface is a device that readsanalog data into the computer from an externalsource: (a) true of (b) false?

38.6. A programmable logic controller normally replaceswhich one of the following in control applications:(a) computer numerical control, (b) distributedprocess control, (c) humans, (d) industrial robots,or (e) relay control panel?

38.7. The standard coordinate system for numerical con-trol machine tools is based on which one of thefollowing: (a) Cartesian coordinates, (b) cylindricalcoordinates, or (c) polar coordinates?

38.8. Identify which of the following applications arepoint-to-point and not continuous path operations(three correct answers): (a) arc welding, (b)

drilling, (c) hole punching in sheet metal, (d) mill-ing, (e) spot welding, and (f) turning?

38.9. The ability of a positioning system to return to apreviously defined location is measured by whichone of the following terms: (a) accuracy, (b) controlresolution, or (c) repeatability?

38.10. TheAPT commandGORGTis which of the follow-ing (two best answers): (a) continuous path com-mand, (b) geometry statement involving a volumeof revolution about a central axis, (c) name of thehumanoid in the latest Star Wars movie, (d) point-to-point command, and (e) tool path command inwhich the tool must go right in the next move?

38.11. The arm-and-body of a robot manipulator gener-ally performs which one of the following functionsin an application: (a) holds the end effector, (b)orients the end effector within the work volume, or(c) positions the wrist within the work volume?

38.12. A SCARA robot is normally associated with whichone of the following applications: (a) arc welding,(b) assembly, (c) inspection, (d) machine loadingand unloading, or (e) resistance welding?

38.13. In robotics, spray-painting applications are classi-fied as which of the following: (a) continuous pathoperation or (b) point-to-point operation?

38.14. Which of the following are characteristics of worksituations that tend to promote the substitution of arobot in place of a human worker (three bestanswers): (a) frequent job changeovers, (b) hazard-ous work environment, (c) repetitive work cycle,(d) multiple work shifts, and (e) task requiresmobility?

PROBLEMS

Open-Loop Positioning Systems

38.1. A leadscrew with a 7.5 mm pitch drives a worktablein a numerical control positioning system. The lead-screw is powered by a stepping motor which has 200step angles. The worktable is programmed to movea distance of 120 mm from its present position at atravel speed of 300 mm/min. Determine (a) thenumber of pulses required to move the table thespecified distance and (b) the required motor speedand pulse rate to achieve the desired table speed.

38.2. Referring to Problem 38.1, the mechanical inaccu-racies in the open-loop positioning system can bedescribed by a normal distribution whose standarddeviation ¼ 0.005 mm. The range of the worktableaxis is 500 mm, and there are 12 bits in the binaryregister used by the digital controller to store the

programmed position. For the positioning system,determine (a) control resolution, (b) accuracy, and(c) repeatability. (d) What is the minimum numberof bits that the binary register should have so thatthe mechanical drive system becomes the limitingcomponent on control resolution?

38.3. A stepping motor has 200 step angles. Its outputshaft is directly coupled to leadscrew with pitch ¼0.250 in. A worktable is driven by the leadscrew.The table must move a distance of 5.00 in from itspresent position at a travel speed of 20.0 in/min.Determine (a) the number of pulses required tomove the table the specified distance and (b) therequired motor speed and pulse rate to achieve thespecified table speed.


E1C38 11/09/2009 18:1:5 Page 915

38.4. A steppingmotor with 100 step angles is coupled to aleadscrew through a gear reduction of 9:1 (9 rotationsof the motor for each rotation of the leadscrew). Theleadscrew has 5 threads/in. The worktable driven bythe leadscrew must move a distance ¼ 10.00 in at afeed rate of 30.0 in/min. Determine (a) number ofpulses required to move the table, and (b) the re-quired motor speed and pulse rate to achieve thedesired table speed.

38.5. The drive unit for a positioning table is driven by aleadscrew directly coupled to the output shaft of astepping motor. The pitch of the leadscrew ¼ 0.18in. The table must have a linear speed ¼ 35 in/min,and a positioning accuracy ¼ 0.001 in. Mechanicalerrors in the motor, leadscrew, and table connec-tion are characterized by a normal distribution withstandard deviation ¼ 0.0002 in. Determine (a) theminimum number of step angles in the steppingmotor to achieve the accuracy, (b) the associatedstep angle, and (c) the frequency of the pulse trainrequired to drive the table at the desired speed.

38.6. The positioning table for a component insertionmachine uses a stepping motor and leadscrewmechanism. The design specifications require atable speed of 40 in/min and an accuracy¼ 0.0008in. The pitch of the leadscrew ¼ 0.2 in, and thegear ratio ¼ 2:1 (two turns of the motor for eachturn of the leadscrew). The mechanical errors inthe motor, gear box, leadscrew, and table con-nection are characterized by a normal distribu-tion with standard deviation ¼ 0.0001 in.Determine (a) the minimum number of stepangles in the stepping motor, and (b) the fre-quency of the pulse train required to drive thetable at the desired maximum speed.

38.7. The drive unit of a positioning table for a compo-nent insertion machine is based on a steppingmotor and leadscrew mechanism. The specifica-tions are for the table speed to be 25 mm/s overa 600 mm range and for the accuracy to be 0.025mm. The pitch of the leadscrew ¼ 4.5 mm, and thegear ratio ¼ 5:1 (five turns of the motor for eachturn of the leadscrew). The mechanical errors inthe motor, gear box, leadscrew, and table connec-tion are characterized by a normal distribution withstandard deviation ¼ 0.005 mm. Determine (a) the

minimum number of step angles in the steppingmotor, and (b) the frequency of the pulse trainrequired to drive the table at the desired maximumspeed for the stepping motor in part (a).

38.8. The two axes of an x-y positioning table are eachdriven by a steppingmotor connected to a leadscrewwith a 10:1 gear reduction. The step angle on eachstepping motor is 7.5�. Each leadscrew has a pitch¼5.0 mm and provides an axis range ¼ 300.0 mm.There are 16 bits in each binary register used by thecontroller to store position data for the two axes.(a) What is the control resolution of each axis?(b) What are the required rotational speeds andcorresponding pulse train frequencies of each step-ping motor in order to drive the table at 600 mm/min in a straight line from point (25,25) to point(100,150)? Ignore acceleration.

38.9. The y-axis of an x-y positioning table is driven by astepping motor that is connected to a leadscrewwith a 3:1 gear reduction (three turns of the motorfor each turn of the leadscrew). The steppingmotorhas 72 step angles. The leadscrew has 5 threads/inand provides an axis range ¼ 30.0 in. There are 16bits in each binary register used by the controller tostore position data for the axis. (a) What is thecontrol resolution of the y-axis? Determine (b) therequired rotational speed of the y-axis steppingmotor and (c) the corresponding pulse train fre-quency to drive the table in a straight line frompoint (x ¼ 20 in, y ¼ 25 in) to point (x ¼ 4.5 in, y ¼7.5 in) in exactly 30 sec. Ignore acceleration.

38.10. The two axes of an x-y positioning table are eachdriven by a stepping motor connected to a lead-screwwith a 4:1 gear reduction. The number of stepangles on each stepping motor is 200. Each lead-screw has a pitch ¼ 5.0 mm and provides an axisrange ¼ 400.0 mm. There are 16 bits in each binaryregister used by the controller to store positiondata for the two axes. (a) What is the controlresolution of each axis? (b) What are the requiredrotational speeds and corresponding pulse trainfrequencies of each stepping motor in order todrive the table at 600 mm/min in a straight linefrom point (25,25) to point (300,150)? Ignoreacceleration.

Closed-Loop Positioning Systems

38.11. A numerical control (NC) machine tool table ispowered by a servomotor, leadscrew, and opticalencoder. The leadscrew has a pitch¼ 5.0 mm and isconnected to the motor shaft with a gear ratio of16:1 (16 turns of the motor for each turn of theleadscrew). The optical encoder is connected

directly to the leadscrew and generates 200 pulses/rev of the leadscrew. The table must move a dis-tance ¼ 100 mm at a feed rate ¼ 500 mm/min.Determine (a) the pulse count received by thecontrol system to verify that the table has movedexactly 100 mm; and (b) the pulse rate and (c)

Problems 915

E1C38 11/09/2009 18:1:5 Page 916

motor speed that correspond to the feed rate of 500mm/min.

38.12. The worktable of a numerical control machine toolis driven by a closed-loop positioning system whichconsists of a servomotor, leadscrew, and opticalencoder. The leadscrew has 4 threads/in and iscoupled directly to the motor shaft (gear ratio ¼1:1). The optical encoder generates 200 pulses permotor revolution. The table has been programmedto move a distance of 7.5 in at a feed rate¼ 20.0 in/min. (a) How many pulses are received by thecontrol system to verify that the table has movedthe programmed distance? What are (b) the pulserate and (c) motor speed that correspond to thespecified feed rate?

38.13. A leadscrew coupled directly to a dc servomotor isused to drive one of the table axes of an NCmillingmachine. The leadscrew has 5 threads/in. The opti-cal encoder attached to the leadscrew emits 100pulses/rev of the leadscrew. The motor rotates at amaximum speed of 800 rev/min. Determine (a) thecontrol resolution of the system, expressed in lineartravel distance of the table axis; (b) the frequencyof the pulse train emitted by the optical encoderwhen the servomotor operates at maximum speed;and (c) the travel speed of the table at the maxi-mum rpm of the motor.

38.14. Solve the previous problem only the servomotor isconnected to the leadscrew through a gear boxwhose reduction ratio ¼ 12:1 (12 revolutions ofthe motor for each revolution of the leadscrew).

38.15. A leadscrew connected directly to a DC servo-motor is the drive system for a positioning table.The leadscrew pitch ¼ 4 mm. The optical encoderattached to the leadscrew emits 250 pulses/rev ofthe leadscrew. Determine (a) the control resolutionof the system, expressed in linear travel distance ofthe table axis, (b) the frequency of the pulse trainemitted by the optical encoder when the servo-motor operates at 14 rev/s, and (c) the travel speedof the table at the operating speed of the motor.

38.16. A milling operation is performed on an NC machin-ing center. Total travel distance ¼ 300 mm in adirection parallel to one of the axes of the worktable.Cutting speed ¼ 1.25 m/s and chip load ¼ 0.05 mm.The endmilling cutter has four teeth and its diameter¼ 20.0 mm. The axis uses a DC servomotor whoseoutput shaft is coupled to a leadscrew with pitch ¼6.0 mm. The feedback sensing device connected to

the leadscrew is an optical encoder that emits 250pulses per revolution. Determine (a) feed rate andtime to complete the cut, and (b) rotational speed ofthe motor and the pulse rate of the encoder at thefeed rate indicated.

38.17. An end milling operation is carried out along astraight line path that is 325mm long. The cut is in adirection parallel to the x-axis on an NCmachiningcenter. Cutting speed ¼ 30 m/min and chip load ¼0.06 mm. The end milling cutter has two teeth andits diameter ¼ 16.0 mm. The x-axis uses a DCservomotor connected directly to a leadscrewwhose pitch ¼ 6.0 mm. The feedback sensingdevice is an optical encoder that emits 400 pulsesper revolution. Determine (a) feed rate and time tocomplete the cut, and (b) rotational speed of themotor and the pulse rate of the encoder at the feedrate indicated.

38.18. A DC servomotor drives the x-axis of a NC millingmachine table. Themotor is coupled to the table leadscrew using a 4:1 gear reduction (four turns of themotor for each turn of the lead screw). The leadscrew pitch ¼ 6.25 mm. An optical encoder is con-nected to the lead screw. The optical encoder emits500 pulses per revolution. To execute a certain pro-grammed instruction, the table must move frompoint (x ¼ 87.5 mm, y ¼ 35.0) to point (x ¼ 25.0mm, y ¼ 180.0 mm) in a straight-line trajectory at afeed rate ¼ 200 mm/min. Determine (a) the controlresolution of the system for the x-axis only, (b) thecorresponding rotational speed of the motor, and (c)frequency of the pulse train emitted by the opticalencoder at the desired feed rate.

38.19. A DC servomotor drives the y-axis of a NC millingmachine table. The motor is coupled to the tablelead screwwith a gear reduction of 2:1 (two turns ofthe motor shaft for each single rotation of the leadscrew). There are 2 threads/cm in the lead screw.An optical encoder is directly connected to the leadscrew (1:1 gear ratio). The optical encoder emits100 pulses per revolution. To execute a certainprogrammed instruction, the table must movefrom point (x ¼ 25.0 mm, y ¼ 28.0) to point (x¼ 155.0 mm, y ¼ 275.0 mm) in a straight-linetrajectory at a feed rate ¼ 200 mm/min. For they-axis only, determine: (a) the control resolution ofthe mechanical system, (b) rotational speed of themotor, and (c) frequency of the pulse train emittedby the optical encoder at the desired feed rate.

Industrial Robotics

38.20. The largest axis of a Cartesian coordinate robot hasa total range of 750 mm. It is driven by pulleysystem capable of a mechanical accuracy ¼ 0.25

mm and repeatability¼ �0.15 mm. Determine theminimum number of bits required in the binaryregister for the axis in the robot’s control memory.


E1C38 11/09/2009 18:1:6 Page 917

38.21. A stepper motor serves as the drive unit for thelinear joint of an industrial robot. The joint musthave an accuracy of 0.25mm. Themotor is attachedto a leadscrew through a 2:1 gear reduction (twoturns of the motor for one turn of the leadscrew).The pitch of the leadscrew is 5.0 mm. The mechan-ical errors in the system (due to backlash of theleadscrew and the gear reducer) can be repre-sented by a normal distribution with standarddeviation ¼ �0.05 mm. Specify the number ofstep angles that the motor must have in order tomeet the accuracy requirement.

38.22. The designer of a polar configuration robot isconsidering a portion of the manipulator consisting

of a rotational joint connected to its output link.The output link is 25 in long and the rotational jointhas a range of 75�. The accuracy of the joint–linkcombination, expressed as a linear measure at theend of the link which results from rotating the joint,is specified as 0.030 in. The mechanical inaccura-cies of the joint result in a repeatability error¼ �0.030� of rotation. It is assumed that the linkis perfectly rigid, so there are no additional errorsdue to deflection. (a) Show that the specifiedaccuracy can be achieved, given the repeatabilityerror. (b) Determine the minimum number of bitsrequired in the binary register of the robot’s con-trol memory to achieve the specified accuracy.

Problems 917

E1C39 11/11/2009 16:54:48 Page 918

39INTEGRATEDMANUFACTURINGSYSTEMS

Chapter Contents

39.1 Material Handling

39.2 Fundamentals of Production Lines39.2.1 Methods of Work Transport39.2.2 Product Variations

39.3 Manual Assembly Lines39.3.1 Cycle Time Analysis39.3.2 Line Balancing and Repositioning

Losses

39.4 Automated Production Lines39.4.1 Types of Automated Lines39.4.2 Analysis of Automated Production

Lines

39.5 Cellular Manufacturing39.5.1 Part Families39.5.2 Machine Cells

39.6 Flexible Manufacturing Systems and Cells39.6.1 Integrating the FMS Components39.6.2 Applications of Flexible

Manufacturing Systems

39.7 Computer Integrated Manufacturing

The manufacturing systems discussed in this chapter consistof multiple workstations and/or machines whose operationsare integrated by means of a material handling subsystemthat moves parts or products between stations. In addition,most of these systemsusecomputer control to coordinate theactions of the stations andmaterial handling equipment andto collect data on overall system performance. Thus, thecomponents of an integrated manufacturing system are(1)workstations and/ormachines, (2)material handling equip-ment,and(3) computer control. In addition, humanworkersare required to manage the system, and workers may beused to operate the individual workstations and machines.

Integratedmanufacturing systems includemanual andautomated production lines, manufacturing cells (fromwhich the term ‘‘cellular manufacturing’’ is derived), andflexiblemanufacturing systems, all of which are described inthis chapter. In the final section we define computer inte-grated manufacturing (CIM), the ultimate integrated man-ufacturing system. Let us begin by providing a conciseoverview of material handling, the physical integrator inintegrated manufacturing systems.

39.1 MATERIAL HANDLING

Material handling is defined as ‘‘the movement, storage,protection and control of materials throughout the manufac-turing and distribution process’’1 The term is usually associ-ated with activities that occur inside a facility, as contrastedwith transportation between facilities that involves rail, truck,air, or waterway delivery of goods.

Materials must be moved during the sequence of man-ufacturing operations that convert them into final product.

1This definition is published each year in the Annual Report of theMaterial Handling Industry of America (MHIA), the trade associationfor material handling companies doing business in North America.

918

E1C39 11/11/2009 16:54:48 Page 919

Materialhandling functions inmanufacturing include (1) loadingandpositioningworkunitsat each workstation, (2) unloading work units from the station, and (3) transporting workunits between workstations. Loading involves moving the work units into the productionmachine froma location in close proximity to orwithin theworkstation. Positioningmeanslocating the work units in a fixed orientation relative to the processing or assemblyoperation. At the end of the operation, the work units are unloaded or removed from thestation. Loading and unloading are accomplishedmanually or by automated devices suchas industrial robots. If the manufacturing operations require multiple workstations, thenthe unitsmust be transported fromone station to the next in the sequence. Inmany cases, atemporary storage function must also be provided by the material handling system, asworkunits await their turn at eachworkstation.Thepurposeof storage in this instance is tomake sure that work is always present at each station, so that idle time of workers andequipment is avoided.

Material handling equipment and methods used in manufacturing can be dividedinto the following general categories: (1) material transport, (2) storage, and (3) unitizing.

Material transport equipment is used to move parts and materials between work-stations in the factory. Thismovementmay include intermediate stops for temporary storageof work-in-process. There are fivemain types ofmaterial transport equipment: (1) industrialtrucks, the most important of which are fork lift trucks, (2) automated guided vehicles,(3) rail-guided vehicles, (4) conveyors, and (5) hoists and cranes. This equipment is brieflydescribed in Table 39.1.

Two general categories of material transport equipment can be distinguished,according to the type of routing betweenworkstations: fixed and variable. In fixed routing,all of thework units aremoved through the same sequence of stations. This implies that theprocessing sequence required on all work units is either identical or very similar. Fixedrouting is used onmanual assembly lines and automated production lines. Typical materialhandling equipment used in fixed routing includes conveyors and rail-guided vehicles. Invariable routing, different work units aremoved through different workstation sequences,meaning that the manufacturing system processes or assembles different types of parts orproducts.Manufacturing cells and flexiblemanufacturing systems usually operate this way.Typical handling equipment found in variable routing includes industrial trucks, automatedguided vehicles, and hoists and cranes.

Storage systems in factories are used for temporary storage of rawmaterials, work-in-process, and finished products. Storage systems can be classified into two general categories:

TABLE 39.1 Five types of material transport equipment.

Type Description Typical Production Applications

Industrial trucks Powered trucks include fork lift trucks as inFigure 39.1(a). Hand trucks includewheeled platforms and dollies

Movement of pallet and container loads infactories and warehouses. Hand trucks usedfor small loads over short distances

Automated guidedvehicles

Independently operated, self-propelledvehicles guided along defined pathways,as in Figure 39.1(b). Powered by on-boardbatteries

Movement of parts and products in assemblylines and flexible manufacturing systems

Rail-guidedvehicles

Motorized vehicles guided by a fixed railsystem. Powered by electrified rail

Monorails used for overhead delivery of largecomponents and subassemblies

Conveyors Apparatus to move items along fixed pathusing chain, moving belt, rollers (Figure39.1(c), or other mechanical drive

Movement of large quantities of items betweenspecific locations. Movement of product onproduction lines

Hoists and cranes Apparatus used for vertical lifting (hoists)and horizontal movement (cranes)

Lifting and transporting heavy materials andloads

Section 39.1/Material Handling 919

E1C39 11/11/2009 16:54:49 Page 920

(1) conventional storage methods and equipment, which include bulk storage in an openarea, rack systems, and shelves; and (2) automated storage systems, which include racksystems served by automatic cranes that store and retrieve pallet loads.

Finally, unitizing refers to containers used to hold individual items during transportand storage, as well as equipment used to make up such unit loads. Containers includepallets, tote pans, boxes, and baskets that hold parts during handling. Unitizing equipmentincludes palletizers that are used to load and stack cartons onto pallets and depalletizersthat are used to accomplish the unloading operation. Palletizers and depalletizers aregenerally associated with cartons of finished product leaving a facility and boxes of rawmaterials coming into the facility, respectively.

In Section 1.4.1, we described four types of plant layout: (1) fixed position layout,(2) process layout, (3) cellular layout, and (4) product layout. In general, different types ofmaterial handling methods and equipment are associated with these four types, assummarized in Table 39.2.

39.2 FUNDAMENTALS OF PRODUCTION LINES

Production lines are an important class of manufacturing system when large quantities ofidentical or similarproducts are tobemade.Theyare suited to situationswhere the totalworkto be performed on the product or part consists of many separate steps. Examples include

TABLE 39.2 Types of material handling methods and systems generally associated with the four types of plantlayout.

Layout Type Features Typical Methods and Equipment

Fixed-position Product is large and heavy, low production rates Cranes, hoists, fork lift trucksProcess Medium and hard product variety, low and

medium production ratesFork lift trucks, automated guided vehicles,manual loading at workstations

Cellular Soft product variety, medium production rates Conveyors, manual handling for loading andmoving between stations

Product No product variety or soft product variety, highproduction rates

Conveyors for product flow, fork lift trucks orautomated guided vehicles to deliver parts tostations

FIGURE 39.1 Severaltypes of materialhandling equipment:

(a) fork lift truck, (b) auto-matedguidedvehicle, and(c) roller conveyor.

Deck forunit loads

BumperDrive

wheels

Rolls

Frame

(c)

(a)

Fork carriage

Forks

Mast

Overhead safetyguard

(b)

920 Chapter 39/Integrated Manufacturing Systems

E1C39 11/11/2009 16:54:49 Page 921

assembled products (e.g., automobiles and appliances) and mass-produced machined partson which multiple machining operations are required (e.g., engine blocks and transmissionhousings). In a production line, the total work is divided into small tasks, and workers ormachines perform these tasks with great efficiency. For purposes of organization we divideproduction lines into twobasic types:manual assembly lines and automatedproduction lines.However, hybrid lines consisting of both manual and automated operations are notuncommon. Before examining these particular systems, let us consider some of the generalissues involved in production line design and operation.

A production line consists of a series of workstations arranged so that the productmoves from one station to the next, and at each location a portion of the total work isperformed on it, as depicted in Figure 39.2. The production rate of the line is limited by itsslowest station.Workstationswhosepace is faster than the slowestwill ultimatelybe limitedby that bottleneck station. Transfer of the product along the line is usually accomplished bya conveyor system or mechanical transfer device, although some manual lines simply passthe product from worker to worker by hand. Production lines are associated with massproduction. If the product quantities are high and the work can be divided into separatetasks that can be assigned to individual workstations, then a production line is the mostappropriate manufacturing system.

39.2.1 METHODS OF WORK TRANSPORT

There are various ways of moving work units from one workstation to the next. The twobasic categories are manual and mechanized.

Manual Methods of Work Transport Manual methods involve passing the workunits between stations by hand. These methods are associated with manual assemblylines. In some cases, the output of each station is collected in a box or tote pan; when thebox is full it is moved to the next station. This can result in a significant amount of in-process inventory, which is undesirable. In other cases, work units are moved individuallyalong a flat table or unpowered conveyor (e.g., a roller conveyor). When the task isfinished at each station, the worker simply pushes the unit toward the downstreamstation. Space is usually allowed for one ormore units to collect between stations, therebyrelaxing the requirement for all workers to perform their respective tasks in sync. Oneproblem associated with manual methods of work transport is the difficulty in controllingthe production rate on the line. Workers tend to work at a slower pace unless somemechanical means of pacing them is provided.

Mechanized Methods of Work Transport Powered mechanical systems are com-monlyused tomoveworkunits along a production line. These systems include lift-and-carrydevices, pick-and-place mechanisms, powered conveyors (e.g., overhead chain conveyors,belt conveyors, and chain-in-floor conveyors), and other material handling equipment,

FIGURE 39.2 General

configuration of aproduction line.

Workpart transport system Partially completed work units

Starting work units

Stations: 1 2 3 n – 1 n

Finished partsor products

Workstations

Section 39.2/Fundamentals of Production Lines 921

E1C39 11/11/2009 16:54:49 Page 922

sometimes combining several types on the same line. Three major types of work transfersystemsareusedonproduction lines: (1) continuous transfer, (2) synchronous transfer, and(3) asynchronous transfer.

Continuous transfer systemsconsistofacontinuouslymovingconveyor thatoperatesataconstantvelocity.Thecontinuoustransfersystemismostcommononmanualassembly lines.Twocasesaredistinguished: (1)parts are fixed to the conveyor and (2)parts canbe removedfromtheconveyor. In the first case, theproduct isusually largeandheavy(e.g., automobile,washing machine) and cannot be removed from the line. The worker must therefore walkalongwith themoving conveyor to complete the assigned task for that unit while it is in thestation. In the second case, the product is small enough that it can be removed from theconveyor to facilitate the work at each station. Some of the pacing benefits are lost in thisarrangement, since each worker is not required to finish the assigned tasks within a fixedtime period. On the other hand, this case allows greater flexibility to each worker to dealwith any technical problems that may be encountered on a particular work unit.

Insynchronoustransfersystems,workunitsaresimultaneouslymovedbetweenstationswith a quick, discontinuous motion. These systems are also known by the name intermittenttransfer,which characterizes the type ofmotion experienced by thework units. Synchronoustransfer includespositioningof theworkat the stations,which is a requirement for automatedlines that use this mode of transfer. Synchronous transfer is not common for manual lines,because the taskateachandeverystationmustbefinishedwithin thecycle timeor theproductwill leave the station as an incomplete unit. This rigid pacing discipline is stressful to humanworkers. By contrast, this type of pacing lends itself to automated operation.

Asynchronous transfer allows each work unit to depart its current station whenprocessing has been completed. Each unit moves independently, rather than synchro-nously. Thus, at any given moment, some units on the line are moving between stations,while others are positioned at stations. Associated with the operation of an asynchronoustransfer system is the tactical use of queues between stations. Small queues of work unitsare permitted to form in front of each station, so that variations inworker task timeswill beaveraged and stations will always have work waiting for them. Asynchronous transfer isused for both manual and automated production systems.

39.2.2 PRODUCT VARIATIONS

Production lines can be designed to cope with variations in productmodels. Three types ofline can be distinguished: (1) single model line, (2) batch model line, and (3) mixed modelline. A single model line is one that produces only onemodel, and there is no variation inthe model. Thus, the tasks performed at each station are the same on all product units.

Batch model and mixed model lines are designed to produce two or more differentproduct models on the same line, but they use different approaches for dealing with themodel variations. As its name suggests, a batchmodel line produces eachmodel in batches.The workstations are set up to produce the desired quantity of the first model; then thestations are reconfigured to produce the desired quantity of the next model; and so on.Production time is lost between batches due to the setup changes. Assembled products areoften made using this approach when the demand for each product is medium and theproduct variety is alsomedium. The economics in this case favor the use of one productionline for several products rather than using many separate lines for each model.

A mixed model line also produces multiple models; however, the models areintermixed on the same line rather than being produced in batches. While a particularmodel is being worked on at one station, a different model is being processed at the nextstation. Each station is equipped with the necessary tools and is sufficiently versatile toperform the variety of tasks needed to produce any model that moves through it. Manyconsumer products are assembledonmixedmodel lineswhen the level of product variety is


E1C39 11/11/2009 16:54:49 Page 923

soft. Prime examples are automobiles and major appliances, which are characterized byvariations in models and options.

39.3 MANUAL ASSEMBLY LINES

The manual assembly line was an important development in integrated manufacturingsystems. It is of global importance today in the manufacture of assembled productsincluding automobiles and trucks, consumer electronic products, appliances, power tools,and other products made in large quantities.

A manual assembly line consists of multiple workstations arranged sequentially, atwhich assembly operations are performed by human workers, as in Figure 39.3. The usualprocedureonamanual linebeginswith ‘‘launching’’ abasepart onto the front endof the line.Awork carrier is often required to hold the part during itsmovement along the line. The basepart travels througheachof the stationswhereworkers performtasks that progressivelybuildtheproduct.Componentsareadded to thebasepart at each station, so thatall taskshavebeencompleted when the product exits the final station. Processes accomplished on manualassembly lines include mechanical fastening operations (Chapter 32), spot welding (Section30.2), hand soldering (Section 31.2), and adhesive joining (Section 31.3).

39.3.1 CYCLE TIME ANALYSIS

Equations can be developed to determine the required number of workers and work-stations on amanual assembly line tomeet a given annual demand. Suppose the problem isto design a single model line to satisfy annual demand for a certain product. Managementmust decide how many shifts per week the line will operate and the number of hours pershift. If we assume 50 weeks per year, then the required hourly production rate of the linewill be given by

Rp ¼ Da

50SwHshð39:1Þ

FIGURE 39.3 A portionof a manual assemblyline. Each worker

performs a task at his/herworkstation. A conveyormoves parts on work

carriers from one stationto the next.

Section 39.3/Manual Assembly Lines 923

E1C39 11/11/2009 16:54:49 Page 924

where Rp ¼ the actual average production rate, units/hr; Da ¼ annual demand for theproduct, units/yr;Sw¼number of shifts/wk; andHsh¼ hr/shift. If the line operates 52weeksrather than 50, thenRp¼Da/52SwHsh. The corresponding average production timeper unitis the reciprocal of Rp

Tp ¼ 60

Rpð39:2Þ

where Tp ¼ actual average production time, converted to minutes.Unfortunately, the line may not be able to operate for the entire time given by 50

SwHsh, because of lost time due to reliability problems. These reliability problems includemechanical and electrical failures, tools wearing out, power outages, and similar malfunc-tions. Accordingly, the line must operate at a faster time than Tp to compensate for theseproblems. IfE¼ line efficiency, which is the proportion of uptimeon the line, then the cycletime of the line Tc is given by

Tc ¼ ETp ¼ 60E

Rpð39:3Þ

Anyproduct contains a certainwork content that represents all of the tasks that are tobe accomplished on the line. This work content requires an amount of time called theworkcontent time Twc. This is the total time required to make the product on the line. If weassume that the work content time can be divided evenly among the workers, so that everyworker has anequalworkloadwhose time toperformequalsTc, then theminimumpossiblenumber of workers wmin in the line can be determined as

wmin ¼ Minimum Integer � Twc

Tcð39:4Þ

If each worker is assigned to a separate workstation, then the number of workstations isequal to the number of workers; that is nmin ¼ wmin.

There are two practical reasons why this minimum number of workers cannot beachieved: (1) imperfect balancing, inwhich someworkers are assigned an amount ofworkthat requires less time thanTc, and this inefficiency increases the total number of workersneeded on the line; and (2) repositioning losses, in which some time is lost at each stationto reposition the work or the worker, so that the service time actually available at eachstation is less than Tc, and this will also increase the number of workers on the line.

39.3.2 LINE BALANCING AND REPOSITIONING LOSSES

One of the biggest technical problems in designing and operating a manual assembly line isline balancing. This is the problemof assigning tasks to individual workers so that all workershaveanequal amountofwork.Recall that theentiretyofwork tobeaccomplishedon the lineis given by the work content. This total work content can be divided intominimum rationalwork elements, each element concerned with adding a component or joining them orperforming some other small portion of the total work content. The notion of a minimumrational work element is that it is the smallest practical amount of work into which the totaljob can be divided. Different work elements require different times, and when they aregrouped into logical tasks and assigned to workers, the task times will not be equal. Thus,simplydue to thevariablenatureof element times, someworkerswill endupwithmorework,while other workers will have less. The cycle time of the assembly line is determined by thestation with the longest task time.

One might think that although the work element times are different, it should bepossible to find groups of elements whose sums (task times) are nearly equal, if not


E1C39 11/11/2009 16:54:49 Page 925

perfectly equal. What makes it difficult to find suitable groups is that there are severalconstraints on this combinatorial problem. First, the linemust be designed to achieve somedesired production rate, which establishes the cycle timeTc at which the linemust operate,as provided by Eq. (39.3). Therefore, the sum of the work element times assigned to eachstation must be less than or equal to Tc.

Second, there are restrictions on the order in which the work elements can beperformed. Some elements must be done before others. For example, a hole must be drilledbefore it can be tapped. A screw that will use the tapped hole to attach a mating componentcannot be fastened before the hole has been drilled and tapped. These kinds of requirementson the work sequence are called precedence constraints. They complicate the line balancingproblem. A certain element that might be allocated to a worker to obtain a task time ¼ Tc

cannot be added because it violates a precedence constraint.These andother limitationsmake it virtually impossible to achieve perfect balancingof

the line, whichmeans that some workers will require more time to complete their tasks thanothers. Methods of solving the line balancing problem, that is, allocating work elements tostations, are discussed in other references—excellent references indeed, such as [10]. Theinability toachieveperfectbalancing results in some idle timeatmost stations.Becauseof thisidle time, theactualnumberofworkers requiredon the linewillbegreater than thenumberofworkstations given by Eq. (39.4).

A measure of the total idle time on a manual assembly line is given by the balancingefficiencyEb, definedas the totalworkcontent timedividedby the total availableservice timeon the line. The total work content time is equal to the sum of the times of all work elementsthat are to be accomplished on the line. The total available service time on the line ¼ wTs,wherew¼numberofworkerson the line; andTs¼ the longest service timeon the line; that is,Ts ¼ Max{Tsi} for i ¼ 1, 2, . . . n, where Tsi ¼ the service time (task time) at station i, min.

The reader may wonder why we are using a new term Ts rather than the previouslydefined cycle time Tc. The reason is that there is another time loss in the operation of aproduction line in addition to idle time from imperfect balancing. Let us call it therepositioning time Tr. It is the time required in each cycle to reposition the worker, orthe work unit, or both. On a continuous transfer line where work units are attached to theline and move at a constant speed, Tr is the time taken by the worker to walk from the unitjust completed to the next unit coming into the station. In all manual assembly lines, therewill be some lost time due to repositioning. We assume that Tr is the same for all workers,although in fact repositioningmay require different times at different stations.We can relateTs, Tc, and Tr as follows:

Tc ¼ Ts þ Tr ð39:5ÞThe definition of balancing efficiencyEb can now be written in equation form as follows:

Eb ¼ Twc

wTsð39:6Þ

A perfect line balance yields a value of Eb ¼ 1.00. Typical line balancing efficiencies inindustry range between 0.90 and 0.95.

Equation (39.6) can be rearranged to obtain the actual number of workers requiredon a manual assembly line:

w ¼ Minimum Integer � Twc

TsEbð39:7Þ

Theutility of this relationship suffers from the fact that the balancing efficiencyEb dependson w in Eq. (39.6). Unfortunately, we have an equation where the thing to be determineddepends on a parameter that, in turn, depends on the thing itself. Notwithstanding thisdrawback, Eq. (39.7) defines the relationship among the parameters in a manual assembly

Section 39.3/Manual Assembly Lines 925

E1C39 11/11/2009 16:54:51 Page 926

line.Usinga typical valueofEbbasedonsimilarprevious lines, it canbeused toestimate thenumber of workers required to produce a given assembly.

Example 39.1Manual AssemblyLine

A manual assembly line is being planned for a product whose annual demand ¼ 90,000units. A continuously moving conveyor will be used with work units attached. Workcontent time ¼ 55 min. The line will run 50 wk/yr, 5 shifts/wk, and 8 hr/day. Each workerwill be assigned to a separate workstation. Based on previous experience, assume lineefficiency ¼ 0.95, balancing efficiency ¼ 0.93, and repositioning time ¼ 9 sec. Determine(a) hourly production rate to meet demand, (b) number of workers and workstationsrequired, and (c) for comparison, the ideal minimum value as given by wmin as given byEq. (39.4).

Solution: (a) Hourly production rate required to meet annual demand is given byEq. (39.1):

Rp ¼ 90; 000

50(5)(8)¼ 45 units=hr

(b) With a line efficiency of 0.95, the ideal cycle time is

Tc ¼ 60(0:95)

45¼ 1:2667min

Given that repositioning time Tr ¼ 9 sec ¼ 0.15 min, the service time is

Ts ¼ 1:2667� 0:150 ¼ 1:1167min

Workers required to operate the line, by Eq. (39.7) equals

w ¼ Minimum Integer � 55

1:1167(0:93)¼ 52:96 ! 53 workers

With one worker per station, n ¼ 53 workstations.(c) This compares with the ideal minimum number of workers given by Eq. (39.4):

wmin ¼ Minimum Integer � 55

1:2667¼ 43:42 ! 44 workers

It is clear that lost time due to repositioning and imperfect line balancing take their toll inthe overall efficiency of a manual assembly line. n

Thenumberofworkstations onamanual assembly linedoes not necessarily equal thenumberofworkers. For largeproducts, itmaybepossible toassignmore thanoneworker toa station. This practice is common in final assembly plants that build cars and trucks. Forexample, two workers in a station might perform assembly tasks on opposite sides of thevehicle. The number of workers in a given station is called the station manning level Mi.Averaging the manning levels over the entire line,

M ¼ w

nð39:8Þ

whereM¼ averagemanning level for the assembly line;w¼ number of workers on the line;and n¼ number of stations. Naturally,w and nmust be integers.Multiplemanning conservesvaluable floor space in the factory because it reduces the number of stations required.

Another factor that affects manning level on an assembly line is the number ofautomated stations on the line, including stations that employ industrial robots (Section38.4). Automation reduces the required labor force on the line, although it increases theneed for technically trained personnel to service andmaintain the automated stations. The


E1C39 11/11/2009 16:54:51 Page 927

automobile industry makes extensive use of robotic workstations to perform spot weldingand spray painting on sheet-metal car bodies. The robots accomplish these operations withgreater repeatability than humanworkers can,which translates into higher product quality.

39.4 AUTOMATED PRODUCTION LINES

Manual assembly lines generally use a mechanized transfer system to move parts betweenworkstations, but the stations themselves are operated by human workers. An automatedproduction line consists of automated workstations connected by a parts transfer systemthat is coordinated with the stations. In the ideal, no human workers are on the line, exceptto perform auxiliary functions such as tool changing, loading and unloading parts at thebeginning and end of the line, and repair and maintenance activities. Modern automatedlines are highly integrated systems, operating under computer control.

Operations performed by automated stations tend to be simpler than those per-formed by humans onmanual lines. The reason is that simpler tasks are easier to automate.Operations that are difficult to automate are those requiring multiple steps, judgment, orhuman sensory capability. Tasks that are easy to automate consist of single work elements,quick actuating motions, and straight-line feed motions as in machining.

39.4.1 TYPES OF AUTOMATED LINES

Automated production lines can be divided into two basic categories: (1) those thatperform processing operations such as machining, and (2) those that perform assemblyoperations. An important type in the processing category is the transfer line.

Transfer Lines and Similar Processing Systems A transfer line consists of asequence of workstations that perform production operations, with automatic transferof work units between stations. Machining is the most common processing operation, asdepicted in Figure 39.4. Automatic transfer systems for sheet metalworking and assemblyare also available. In the case of machining, the workpiece typically starts as a metalcasting or forging, and a series of machining operations are performed to accomplish thehigh-precision details (e.g., holes, threads, and finished flat surfaces).

FIGURE 39.4A machining transfer line,

an important type ofautomated productionline.

Section 39.4/Automated Production Lines 927

E1C39 11/11/2009 16:54:52 Page 928

Transfer lines are usually expensivepieces of equipment, sometimes costingmillions ofdollars; they aredesigned forhighpart quantities. The amount ofmachining accomplishedonthe workpart is often significant, but since the work is divided among many stations,production rates are high andunit costs are lowcompared to alternative productionmethods.Synchronous transfer of work units between stations is commonly used on automatedmachining lines.

Avariation of the automated transfer line is the dial indexing machine, Figure 39.5,in which workstations are arranged around a circular worktable, called a dial. The work-table is actuated by a mechanism that provides partial rotation of the table on each workcycle. The number of rotational positions is designed tomatch the number of workstationsaround the periphery of the table.Although the configuration of a dial-indexingmachine isquite different from a transfer line, its operation and application are quite similar.

Automated Assembly Systems Automated assembly systems consist of one or moreworkstations that performassembly operations, such as adding components and/or affixingthem to thework unit. Automated assembly systems can be divided into single station cellsand multiple station systems. Single station assembly cells are often organized around anindustrial robot that has been programmed to perform a sequence of assembly steps. Therobot cannot work as fast as a series of specialized automatic stations, so single station cellsare used for jobs in the medium production range.

Multiple station assembly systems are appropriate for high production. They arewidely used formass productionof small products such as ball-point pens, cigarette lighters,flashlights, and similar items consisting of a limited number of components. The number ofcomponents and assembly steps is limited because system reliability decreases rapidly withincreasing complexity.

Multiple station assembly systems are available in several configurations, pictured inFigure39.6:(a)in-line,(b)rotary,and(c)carousel.Thein-lineconfigurationistheconventional

FIGURE 39.5Configuration of a

dial-indexing machine.

Workstations

Workstations

Rotarytransfertable

In-line transfervc

(a) (b) (c)Carousel

FIGURE 39.6 Three common configurations of multiple station assembly systems: (a) in-line, (b) rotary,and (c) carousel.


E1C39 11/11/2009 16:54:53 Page 929

transfer line adapted to perform assembly work. These systems are not as massive as theirmachining counterparts. Rotary systems are usually implemented as dial indexingmachines.Carouselassemblysystemsarearrangedasaloop.Theycanbedesignedwithagreaternumberofworkstations thana rotary system.Owing to the loopconfiguration, the carousel allows theworkcarriers tobeautomatically returned to the startingpoint for reuse, anadvantage sharedwith rotary systems but not with transfer lines unless provision for their return is made in thedesign.

39.4.2 ANALYSIS OF AUTOMATED PRODUCTION LINES

Line balancing is a problem on an automated line, just as it is on a manual assembly line.The total work content must be allocated to individual workstations. However, since thetasks assigned to automated stations are generally simpler, and the line often containsfewer stations, the problem of defining what work should be done at each station is not asdifficult for an automated line as for a manual line.

Amoresignificantprobleminautomated lines is reliability.The lineconsistsofmultiplestations, interconnected by a work transfer system. It operates as an integrated system, andwhen one station malfunctions, the entire system is adversely affected. To analyze theoperation of an automated production line, let us assume a system that performs processingoperations and uses synchronous transfer. This model includes transfer lines as well as dialindexing machines. It does not include automated assembly systems, which require anadaptationof themodel [10].Our terminologywillborrowsymbolsfromthefirst twosections:n¼ number of workstations on the line; Tc¼ ideal cycle time on the line; Tr¼ repositioningtime, called the transfer time in a transfer line; andTsi¼ the service time at station i. The idealcycle time Tc is the service time (processing time) for the slowest station on the line plus thetransfer time; that is,

Tc ¼ Tr þMaxfTsig ð39:9ÞIn the operation of a transfer line, periodic breakdowns cause downtime on the entire

line. Let F ¼ frequency with which breakdowns occur, causing a line stoppage; and Td ¼average time the line is downwhen a breakdown occurs. The downtime includes the time forthe repair crew to swing intoaction, diagnose the causeof the failure, fix it, and restart the line.

Based on these definitions, we can formulate the following expression for the actualaverage production time Tp:

Tp ¼ Tc þ FTd ð39:10Þ

whereF¼ downtime frequency, line stops/cycle; andTd¼ downtime inminutes per line stop.Thus, FTd¼ average downtime per cycle. The actual average production rateRp¼ 60/Tp, aspreviously given in Eq. (39.2). It is of interest to compare this rate with the ideal productionrate given by

Rc ¼ 60

Tcð39:11Þ

where Rp and Rc are expressed in pc/hr, given that Tp and Tc are expressed in minutes.Based on these definitions, we can define the line efficiency E for a transfer line. In

the context of automated production systems,E refers to the proportion of uptime on theline and is more a measure of reliability than efficiency:

E ¼ Tc

Tc þ FTdð39:12Þ

Section 39.4/Automated Production Lines 929

E1C39 11/11/2009 16:54:54 Page 930

This is the same relationship as earlier Eq. (39.3), since Tp¼ Tcþ FTd. It should be notedthat the same definition of line efficiency applies to manual assembly lines, except thattechnological breakdowns are not as much of a problem onmanual lines (human workersare more reliable than electromechanical equipment, at least in the sense we arediscussing here).

Line downtime is usually associated with failures at individual workstations.Reasons for downtime include scheduled and unscheduled tool changes, mechanicaland electrical malfunctions, hydraulic failures, and normal equipment wear. Let pi ¼probability or frequency of a failure at station i, then

F ¼Xn

i�1

pi ð39:13Þ

If all pi are assumed equal, or an average value of pi is computed, in either case calling it p,then

F ¼ np ð39:14ÞBoth of these equations clearly indicate that the frequency of line stops increases with thenumberof stationson the line. Stated anotherway, reliability of the linedecreases asweaddmore stations.

Example 39.2AutomatedTransfer Line

Anautomated transfer line has 20 stations and an ideal cycle time of 1.0min. Probability ofa station failure is p¼ 0.01, and the average downtimewhen a breakdown occurs is 10min.Determine (a) average production rate Rp and (b) line efficiency E.

Solution: The frequency of breakdowns on the line is given by F¼ pn¼ 0.01(20)¼ 0.20.The actual average production time is therefore

Tp ¼ 1:0þ 0:20(10) ¼ 3:0 min

(a) Production rate is therefore

Rp ¼ 60

Tp¼ 60

3:0¼ 20 pc/hr

Note that this is far lower than the ideal production rate:

Rc ¼ 60

Tc¼ 60

1:0¼ 60 pc/hr

(b) Line efficiency is computed as

E ¼ Tc

Tp¼ 1:0

3:0¼ 0:333(or 33:3%)

Fromthis examplewe see that ifaproduction lineoperates like this, it spendsmore timedownthan up. Achieving high efficiencies is a real problem in automated production lines. n

The cost of operating an automated production line is the investment cost of theequipment and installation, plus the cost of maintenance, utilities, and labor assignedto the line. These costs are converted to an equivalent uniform annual cost anddivided by the number of hours of operation per year to provide an hourly rate. Thishourly cost rate can be used to figure the unit cost of processing a workpart on theline

Cp ¼ CoTp

60ð39:15Þ


E1C39 11/11/2009 16:54:54 Page 931

whereCp ¼ unit processing cost, $/part; Co¼ hourly rate of operating the line, as definedabove, $/hr; Tp ¼ actual average production time per workpart, min/part; and theconstant 60 converts the hourly cost rate to $/min for consistency of units.

39.5 CELLULAR MANUFACTURING

Cellular manufacturing refers to the use of work cells that specialize in the production offamilies of parts or products made in medium quantities. Parts (and products) in thisquantity range are traditionally made in batches, and batch production requires downtimefor setupchangeovers andhashigh inventorycarrying costs.Cellularmanufacturing isbasedon an approach called group technology (GT), whichminimizes the disadvantages of batchproductionby recognizing thatalthough thepartsaredifferent, theyalsopossess similarities.When these similarities are exploited in production, operating efficiencies are improved.The improvement is typically achieved by organizing the production aroundmanufacturingcells. Each cell is designed to produce one part family (or a limited number of part families),thereby following the principle of specialization of operations. The cell includes specialproduction equipment and custom-designed tools and fixtures, so that the production of thepart families can be optimized. In effect, each cell becomes a factory within the factory.

39.5.1 PART FAMILIES

A central feature of cellular manufacturing and group technology is the part family. A partfamily is a group of parts that possess similarities in geometric shape and size, or in theprocessing steps used in theirmanufacture. It is not unusual for a factory that produces 10,000different parts to be able to groupmost of those parts into 20 to 30 part families. In each partfamilytheprocessingstepsaresimilar.Therearealwaysdifferencesamongpartsinafamily,butthesimilaritiesarecloseenoughthatthepartscanbegroupedintothesamefamily.Figures39.7and39.8showtwodifferentpart families.ThepartsshowninFigure39.7havethesamesizeandshape; however, their processing requirements are quite different because of differences inwork material, production quantities, and design tolerances. Figure 39.8 shows several partswith geometries that differ, but their manufacturing requirements are quite similar.

There are several ways by which part families are identified in industry. One methodinvolves visual inspectionof all thepartsmade in the factory (orphotosof theparts) andusingbest judgment to group them into appropriate families.Another approach, calledproductionflowanalysis,uses information contained on route sheets (Section 40.1.1) to classify parts. Ineffect, parts with similar manufacturing steps are grouped into the same family.

A third method, usually the most expensive but most useful, is parts classification andcoding. Parts classification and coding involve the identification of similarities and differ-ences among parts and relating these parts by means of a numerical coding scheme. Mostclassification and coding systems are one of the following: (1) systems based on part design

FIGURE 39.7 Two partsthat are identical in shape

and size but quitedifferent in manufacturing:(a) 1,000,000 units/yr, tol-

erance=�0.010 in,1015CRsteel, nickel plate; and (b)100/yr, tolerance = �0.001

in, 18-8 stainless steel.

Section 39.5/Cellular Manufacturing 931

E1C39 11/11/2009 16:54:55 Page 932

attributes, (2) systems based on part manufacturing attributes, and (3) systems based onboth design and manufacturing attributes. Common part design and manufacturingattributes used in GT systems are presented in Table 39.3. Because each companyproduces a unique set of parts and products, a classification and coding system thatmay be satisfactory for one company is not necessarily appropriate for another company.Each companymust design its own coding scheme. Parts classification and coding systemsare described more thoroughly in several of our references [8], [10], [11].

Benefits often cited for a well-designed classification and coding system include(1) facilitates formation of part families, (2) permits quick retrieval of part designdrawings, (3) reduces design duplication because similar or identical part designs canbe retrieved and reused rather than designed from scratch, (4) promotes design standard-ization, (5) improves cost estimating and cost accounting, (6) facilitates numerical control(NC) part programming by allowing new parts to use the same basic part program asexisting parts in the same family, (7) allows sharing of tools and fixtures, and (8) aidscomputer-aided process planning (Section 40.1.3) because standard process plans can becorrelated to part family code numbers, so that existing process plans can be reused oredited for new parts in the same family.

FIGURE 39.8 Ten parts

that are different in sizeand shape, but quitesimilar in terms of

manufacturing. All partsare machined fromcylindrical stock by

turning; some partsrequire drilling and/ormilling.

TABLE 39.3 Design and manufacturing attributes typically included in a partsclassification and coding system.

Part Design Attributes Part Manufacturing Attributes

Major dimensions Material type Major process Major dimensionsBasic external shape Part function Operation sequence Basic external shapeBasic internal shape Tolerances Batch size Length/diameter ratioLength/diameter ratio Surface finish Annual production Material type

Machine tools TolerancesCutting tools Surface finish


E1C39 11/11/2009 16:54:55 Page 933

39.5.2 MACHINE CELLS

To fully exploit the similarities among parts in a family, production should be organized usingmachine cells designed to specialize inmaking those particular parts.One of the principles indesigning a group technology machine cell is the composite part concept.

Composite Part Concept Members of a part family possess similar design and/ormanufacturing features. There is usually a correlation between part design features andthemanufacturing operations that produce those features. Round holes aremade by drilling;cylindrical shapes are made by turning; and so on.

Thecompositepart foragivenfamily (not tobeconfusedwithapartmadeofcompositematerial) is a hypothetical part that includes all of the design andmanufacturing attributes ofthe family. In general, an individual part in the family will have some of the features thatcharacterize the family, but not all of them. A production cell designed for the part familywould include those machines required to make the composite part. Such a cell would becapable of producing any member of the family, simply by omitting those operationscorrespondingto featuresnotpossessedbytheparticularpart.Thecellwouldalsobedesignedto allow for size variations within the family as well as feature variations.

To illustrate, consider the composite part in Figure 39.9(a). It represents a family ofrotational parts with features defined in part (b) of the figure.Associatedwith each featureis a certain machining operation, as summarized in Table 39.4. A machine cell to produce

FIGURE 39.9 Composite part concept: (a) the composite part for a family of machined rotational parts,and (b) the individual features of the composite part.

TABLE 39.4 Design features of the composite part in Figure 39.3and the manufacturing operations required to shape those features.

Label Design FeatureCorresponding Manufacturing

Operation

1 External cylinder Turning2 Face of cylinder Facing3 Cylindrical step Turning4 Smooth surface External cylindrical grinding5 Axial hole Drilling6 Counterbore Bore, counterbore7 Internal threads Tapping

Section 39.5/Cellular Manufacturing 933

E1C39 11/11/2009 16:54:56 Page 934

this part family would be designedwith the capability to accomplish all of the operations inthe last column of the table.

Machine Cell Designs Machine cells can be classified according to number ofmachinesand level of automation.Thepossibilities are (a) singlemachine, (b)multiplemachineswithmanual handling, (c) multiple machines with mechanized handling, (d) flexible manu-facturing cell, or (e) flexiblemanufacturing system. These production cells are depicted inFigure 39.10.

The singlemachine cellhasonemachine that ismanually operated.The cellwould alsoinclude fixtures and tools to allow for feature and size variations within the part familyproduced by the cell. The machine cell required for the part family of Figure 39.9 wouldprobably be of this type.

Multiplemachine cellshave twoormoremanually operatedmachines. These cells aredistinguishedby themethodofworkpart handling in the cell,manual ormechanized.Manualhandling means that parts are moved within the cell by workers, usually the machineoperators. Mechanized handling refers to conveyorized transfer of parts from one machineto thenext.Thismaybe requiredby the size andweight of thepartsmade in the cell, or simplyto increaseproduction rate.Our sketchdepicts thework flowasbeinga line; other layouts arealso possible, such as U-shaped or loop.

Flexible manufacturing cells and flexible manufacturing systems consist of auto-mated machines with automated handling. Given the special nature of these integratedmanufacturing systems and their importance, we devote Section 39.6 to their discussion.

FIGURE 39.10 Types ofgroup technologymachine cells: (a) single

machine, (b) multiplemachines with manualhandling, (c) multiple

machines with mecha-nized handling, (d) flexiblemanufacturing cell, and

(e) flexible manufacturingsystem. Key: Man ¼manual operation; Aut ¼automated station.


E1C39 11/11/2009 16:54:56 Page 935

Benefits and Problems in Group Technology The use of machine cells and grouptechnology provide substantial benefits to companies that have the discipline and perse-verance to implement it. The potential benefits include the following: (1) GT promotesstandardization of tooling, fixturing, and setups; (2) material handling is reduced becauseparts are moved within a machine cell rather than the entire factory; (3) productionscheduling is simplified; (4) manufacturing lead time is reduced; (5) work-in-process isreduced; (6) process planning is simpler; (7) worker satisfaction usually improves workingin a cell; and (8) higher quality work is accomplished.

There are several problems in implementing machine cells, however. One obviousproblemisrearrangingproductionmachinesintheplantintotheappropriatemachinecells.Ittakes time to plan and accomplish this rearrangement, and the machines are not producingduring the changeover. The biggest problem in starting aGTprogram is identifying the partfamilies. If the plant makes 10,000 different parts, reviewing all of the part drawings andgrouping the parts into families are substantial tasks that consume a significant amountof time.

39.6 FLEXIBLE MANUFACTURING SYSTEMS AND CELLS

Aflexiblemanufacturing system (FMS) is a highly automatedGTmachine cell, consisting ofa group of processing stations (usually computer numerical control [CNC] machine tools),interconnected by an automatedmaterial handling and storage system, and controlled by anintegrated computer system.AnFMS is capableof processingavarietyofdifferentpart stylessimultaneously under NC program control at the different workstations.

An FMS relies on the principles of group technology. No manufacturing system canbe completely flexible. It cannot produce an infinite range of parts or products. There arelimits to how much flexibility can be incorporated into an FMS. Accordingly, a flexiblemanufacturing system is designed to produce parts (or products) within a range of styles,sizes, and processes. In otherwords, anFMS is capable of producing a single part family or alimited range of part families.

Flexible manufacturing systems vary in terms of number of machine tools and level offlexibility. When the system has only a few machines, the term flexible manufacturing cell(FMC) is sometimes used. Both cell and system are highly automated and computercontrolled. The difference between an FMS and an FMC is not always clear, but it issometimes based on the number of machines (workstations) included. The flexible manu-facturing system consists of four or more machines, while a flexible manufacturing cellconsists of three or fewer machines [10].

To qualify as being flexible, amanufacturing system should satisfy several criteria. Thetests of flexibility in an automated production system are the capability to (1) processdifferent part styles in a nonbatch mode, (2) accept changes in production schedule,(3) respond gracefully to equipment malfunctions and breakdowns in the system, and(4) accommodate the introduction of new part designs. These capabilities are madepossible by the use of a central computer that controls and coordinates the components ofthe system. The most important criteria are (1) and (2); criteria (3) and (4) are softer andcan be implemented at various levels of sophistication.

39.6.1 INTEGRATING THE FMS COMPONENTS

An FMS consists of hardware and software that must be integrated into an efficient andreliable unit. It also includes human personnel. In this section we examine thesecomponents and how they are integrated.

Section 39.6/Flexible Manufacturing Systems and cells 935

E1C39 11/11/2009 16:54:56 Page 936

Hardware Components FMS hardware includes workstations, material handlingsystem, and central control computer. The workstations are CNCmachines in amachiningtype system, plus inspection stations, parts cleaning and other stations, as required. Acentral chip conveyor system is often installed below floor level.

Thematerial handling system is themeans bywhich parts aremoved between stations.The material handling system usually includes a limited capability to store parts. Handlingsystems suitable for automated manufacturing include roller conveyors, automated guidedvehicles,andindustrialrobots.Themostappropriatetypedependsonpartsizeandgeometry,aswell as factors relating to economics and compatibility with other FMS components. Non-rotational parts are often moved in a FMS on pallet fixtures, so the pallets are designed forthe particular handling system, and the fixtures are designed to accommodate the variouspart geometries in the family. Rotational parts are often handled by robots, if weight is not alimiting factor.

The handling system establishes the basic layout of the FMS. Five layout types can bedistinguished: (1) in-line, (2) loop, (3) ladder, (4) open field, and (5) robot-centered cell.Types 1, 3, 4, and 5 are shown in Figure 39.11. Type 2 is shown in Figure 39.10(e). The in-line layout uses a linear transfer system to move parts between processing stations andload/unload station(s). The in-line transfer system is usually capable of two-directionalmovement; if not, then the FMS operates much like a transfer line, and the different partstyles made on the systemmust follow the same basic processing sequence due to the one-direction flow. The loop layout consists of a conveyor loop with workstations locatedaround its periphery. This configuration permits any processing sequence, because anystation is accessible from any other station. This is also true for the ladder layout, in whichworkstations are located on the rungs of the ladder. The open field layout is the mostcomplex FMS configuration, and consists of several loops tied together. Finally, the robot-centered cell consists of a robot whose work volume includes the load/unload positions ofthe machines in the cell.

The FMS also includes a central computer that is interfaced to the other hardwarecomponents. In addition to the central computer, the individual machines and othercomponents generally havemicrocomputers as their individual control units. The functionof the central computer is to coordinate the activities of the components so as to achieve asmooth overall operation of the system. It accomplishes this function bymeans of software.

FMS Software and Control Functions FMS software consists of modules associatedwith the various functions performed by the manufacturing system. For example, onefunction involves downloading NC part programs to the individual machine tools; anotherfunction is concerned with controlling the material handling system; another is concernedwith toolmanagement; and so on. Table 39.5 lists the functions included in the operation ofa typical FMS.Associatedwith each function is one ormore softwaremodules. Terms otherthan those in our table may be used in a given installation. The functions and modules arelargely application specific.

Human Labor An additional component in the operation of a flexible manufacturingsystemor cell is human labor.Duties performed by humanworkers include (1) loading andunloading parts from the system, (2) changing and setting cutting tools, (3) maintenanceand repair of equipment, (4) NC part programming, (5) programming and operating thecomputer system, and (6) overall management of the system.

39.6.2 APPLICATIONS OF FLEXIBLE MANUFACTURING SYSTEMS

Flexible manufacturing systems are typically used for midvolume, midvariety production.If the part or product ismade in high quantities with no style variations, then a transfer line


E1C39 11/11/2009 16:54:56 Page 937

FIGURE 39.11 Four ofthe five FMS layout types:(a) in-line, (b) ladder,

(c) open field, and (d)robot-centered cell. Key:Aut ¼ automated station;

L/UL ¼ load/unloadstation; Insp ¼ inspectionstation; AGV¼ automated

guided vehicle; AGVS ¼automated guided vehiclesystem.

Conveyor

Conveyor

AGVS guidewireAGV

Machines

Machine

Machine

Inspectionstation

Parts in

Parts in/out

Parts in

Parts out

Parts out

v

v

v

v

v

v

(a)

(b)

(c)

(d)

Aut. Aut.

Aut.

Aut. Aut.

Aut. Aut.

AGV

Aut.

Insp.

Aut. Aut. Aut.

Aut. Aut. Aut.

L/UL

L/UL

Section 39.6/Flexible Manufacturing Systems and cells 937

E1C39 11/11/2009 16:54:56 Page 938

or similar dedicatedmanufacturing system ismost appropriate. If the parts are low volumewith high variety, then a stand-alone NCmachine or evenmanual methods would bemoreappropriate. These application characteristics are summarized in Figure 39.12.

Flexible machining systems comprise the most common application of FMS technol-ogy. Owing to the inherent flexibilities and capabilities of computer numerical control, it ispossible to connect several CNC machine tools to a small central computer, and to deviseautomatedmaterial handlingmethods for transferring parts betweenmachines. Figure 39.13shows a flexible machining system consisting of five CNC machining centers and an in-linetransfer system to pick up parts from a central load/unload station and move them to theappropriate machining stations.

In addition tomachining systems, other types of flexiblemanufacturing systems havealso been developed, although the state of technology in these other processes has not

TABLE 39.5 Typical computer functions implemented by application software modules in a flexiblemanufacturing system.

Function Description

NC part programming Development of NC programs for new parts introduced into the system. This includes alanguage package such as APT

Production control Product mix, machine scheduling, and other planning functionsNC program download Part program commands must be downloaded to individual stations from the central

computerMachine control Individual workstations require controls, usually computer numerical controlWorkpart control Monitor status of each workpart in the system, status of pallet fixtures, orders on loading/

unloading pallet fixturesTool management Functions include tool inventory control, tool status relative to expected tool life, tool

changing and resharpening, and transport to and from tool grindingTransport control Scheduling and control of workpart handling systemSystem management Compiles management reports on performance (utilization, piece counts, production

rates, etc.). FMS simulation sometimes included

NC, numerical control; APT, automatically programmed tool; FMS, flexible manufacturing system.

FIGURE 39.12Applicationcharacteristics of flexible

manufacturing systemsand cells relative to othertypes of manufacturing

systems.


E1C39 11/11/2009 16:54:56 Page 939

permitted the rapid implementation that has occurred in machining. The other types ofsystems include assembly, inspection, sheet-metal processing (punching, shearing, bending,and forming), and forging.

Most of the experience in flexible manufacturing systems has been gained inmachining applications. For flexible machining systems, the benefits usually given are(1) highermachine utilization than a conventionalmachine shop—relative utilizations are40% to 50% for conventional batch-type operations and about 75% for a FMS due tobetter work handling, off-line setups, and improved scheduling; (2) reduced work-in-process due to continuous production rather than batch production; (3) lower manufac-turing lead times; and (4) greater flexibility in production scheduling.

39.7 COMPUTER INTEGRATED MANUFACTURING

Distributed computer networks are widely used inmodernmanufacturing plants tomonitorand/or control the integrated systems described in this chapter. Even though some of theoperations are manually accomplished (e.g., manual assembly lines and manned cells),computer systems are utilized for production scheduling, data collection, record keeping,performance tracking, and other information-related functions. In the more automatedsystems (e.g., transfer lines and flexible manufacturing cells), computers directly control theoperations. The term computer integrated manufacturing refers to the pervasive use ofcomputer systems throughout the organization, not only to monitor and control theoperations, but also to design the product, plan themanufacturing processes, and accomplishthe business functions related to production. One might say that CIM is the ultimateintegratedmanufacturing system. In this final section of Part X, we outline the scope of CIMand provide a bridge to Part XI on manufacturing support systems.

FIGURE 39.13 A five-

station flexiblemanufacturing system.(Photo courtesy of

Cincinnati Milacron,Batavia, Ohio.)

Section 39.7/Computer Integrated Manufacturing 939

E1C39 11/11/2009 16:54:57 Page 940

To begin, let us identify four general functions that have to be accomplished in mostmanufacturingenterprises: (1) productdesign, (2)manufacturingplanning, (3)manufactur-ing control, and (4) business functions. Product design is usually an iterative process thatincludes recognition of a need for a product, problem definition, creative synthesis of asolution, analysis andoptimization, evaluation, anddocumentation.Theoverallqualityof theresulting design is likely to be themost important factor upon which the commercial successof a product depends. In addition, a very significant portion of the final cost of the product isdetermined by decisions made during product design. Manufacturing planning is concernedwith converting the engineering drawings and specifications that define the product designinto a plan for producing the product. Manufacturing planning includes decisions on whichparts will be purchased (the ‘‘make-or-buy decision’’), how each ‘‘make’’ part will beproduced, the equipment that will be used, how the work will be scheduled, and so on.Most of these decisions are discussed in Chapter 40 on manufacturing engineering andChapter 41 on production planning. Manufacturing control includes not only control of theindividual processes and equipment in the plant, but also the supporting functions such asshop floor control and quality control, discussed in Chapters 41 and 42, respectively. Finally,the business functions include order entry, cost accounting, payroll, customer billing, andother business-oriented information activities related to manufacturing.

Computer systems play an important role in these four general functions, and theirintegration within the organization is a distinguishing feature of computer integratedmanufacturing, as depicted inFigure 39.14.Computer systems associatedwith product designare called CAD systems (for computer-aided design). Design systems and software includegeometric modeling, engineering analysis packages such as finite element modeling, designreview and evaluation, and automated drafting. Computer systems that support manufactur-ing planning are called CAM systems (for computer-aided manufacturing) and includecomputer-aided process planning, NC part programming, production scheduling, and plan-ning packages such as manufacturing resource planning (discussed in Chapter 41). Manu-facturing control systems include those used in process control, shop floor control, inventorycontrol, and computer-aided inspection for quality control. And computerized businesssystems are used for order entry, customer billing, and other business functions. Customer

(1) (2) (3) (4)

CIM

CADGeometric modelingEngineering analysisDesign review/eval.Automated drafting

Customer feedback to design

CAPPNC part program.Production scheduleMfg resource planning

ComputerizedBusiness Systems

Order entryCustomer billingPayrollAccounting, etc.

Process controlQuality controlShop floor controlInventory control

CAM CAM

Productdesign

Manufacturingplanning

Manufacturingcontrol

Businessfunctions

Customermarket

Factoryoperations

FIGURE 39.14 Four general functions in a manufacturing organization and how computer integrated

manufacturing systems support these functions.


E1C39 11/11/2009 16:54:57 Page 941

orders are entered by the company’s sales force or by the customers themselves into thecomputerized order entry system. The orders include product specifications that provide theinputs to the design department. Based on these inputs, new products are designed on thecompany’s CAD system. The design details serve as inputs to themanufacturing engineeringgroup, where computer-aided process planning, computer-aided tool design, and relatedactivities are performed in advance of actual production. The output from manufacturingengineering provides much of the input data required for manufacturing resource planningand production scheduling. Thus, computer integrated manufacturing provides the informa-tion flows required to accomplish the actual production of the product.

Today, computer integrated manufacturing is implemented in many companies usingenterprise resource planning (ERP), an extension of manufacturing resource planning thatorganizesandintegratestheinformationflowsinacompanythroughasingle,centraldatabase.The functions covered by ERP spread well beyond manufacturing operations; they includesales, marketing, purchasing, logistics, distribution, inventory control, finance, and humanresources. ERP users within a company access and interact with the system using personalcomputers at their own workplaces, whether they are located in offices or in the factory.

REFERENCES

[1] Black, J. T. The Design of the Factory with a Future.McGraw-Hill, New York, 1990.

[2] Black, J. T.‘‘An Overview of Cellular Manufactur-ing Systems and Comparison to Conventional Sys-tems,’’ Industrial Engineering, November 1983,pp. 36–84.

[3] Boothroyd, G., Poli, C., and Murch, L. E. AutomaticAssembly. Marcel Dekker, New York, 1982.

[4] Buzacott, J. A.‘‘Prediction of the Efficiency of Pro-duction Systems without Internal Storage,’’ Interna-tional Journal of Production Research,Vol. 6, No. 3,1968, pp. 173–188.

[5] Buzacott, J. A., and Shanthikumar, J. G. StochasticModels of Manufacturing Systems. Prentice-Hall,Upper Saddle River, New Jersey, 1993.

[6] Chang, T-C., Wysk, R. A., and Wang, H-P. Com-puter-Aided Manufacturing, 3rd ed. Prentice-Hall,Upper Saddle River, New Jersey, 2005.

[7] Chow,W-M.Assembly LineDesign.Marcel Dekker,New York, 1990.

[8] Gallagher, C. C., and Knight, W. A.Group Technol-ogy. Butterworth & Co., Ltd., London, 1973.

[9] Groover, M. P.‘‘Analyzing Automatic TransferLines,’’ Industrial Engineering, Vol. 7, No. 11,1975, pp. 26–31.

[10] Groover, M. P.Automation, Production Systems, andComputer Integrated Manufacturing, 3rd ed. PearsonPrentice-Hall, Upper Saddle River, New Jersey, 2008.

[11] Ham, I., Hitomi, K., and Yoshida, T. Group Tech-nology. Kluwer Nijhoff Publishers, Hingham, Mas-sachusetts, 1985.

[12] Houtzeel, A.‘‘TheMany Faces of Group Technology,’’American Machinist, January 1979, pp. 115–120.

[13] Luggen, W. W. Flexible Manufacturing Cells andSystems. Prentice Hall, Inc., Englewood Cliffs, NewJersey, 1991.

[14] Maleki, R. A. Flexible Manufacturing Systems: TheTechnology and Management. Prentice Hall, Inc.,Englewood Cliffs, New Jersey, 1991.

[15] Moodie, C., Uzsoy, R., and Yih, Y. ManufacturingCells: A Systems Engineering View. Taylor &Francis, Ltd., London, 1995.

[16] Parsai, H., Leep, H., and Jeon, G. The Principles ofGroup Technology and Cellular Manufacturing.John Wiley & Sons, Hoboken, New Jersey, 2006.

[17] Riley, F. J. Assembly Automation, A. ManagementHandbook, 2nd ed. Industrial Press, New York,1999.

[18] Weber, A.‘‘Is Flexibility a Myth?’’ Assembly, May2004, pp. 50–59.

REVIEW QUESTIONS

39.1. What are the main components of an integratedmanufacturing system?

39.2. What are the principal material handling functionsin manufacturing?

39.3. Name the five main types of material transportequipment.

39.4. What is the difference between fixed routing andvariable routing in material transport systems?

Review Questions 941

E1C39 11/11/2009 16:54:58 Page 942

39.5. What is a production line?39.6. What are the advantages of a mixed model line

over a batch model line for producing differentproduct styles?

39.7. What are some of the limitations of a mixed modelline compared to a batch model line?

39.8. Describe how manual methods are used tomove parts between workstations on a produc-tion line.

39.9. Briefly define the three types of mechanized work-part transfer systems used in production lines.

39.10. Why are parts sometimes fixed to the conveyor in acontinuous transfer system in manual assembly?

39.11. Why must a production line be paced at a ratehigher than that required to satisfy the demand forthe product?

39.12. Repositioning time on a synchronous transfer lineis known by a different name; what is that name?

39.13. Why are single station assembly cells generally notsuited to high-production jobs?

39.14. What are some of the reasons for downtime on amachining transfer line?

39.15. Define group technology.39.16. What is a part family?39.17. Define cellular manufacturing.39.18. What is the composite part concept in group

technology?39.19. What is a flexible manufacturing system?39.20. What are the criteria that shouldbe satisfied tomake

an automated manufacturing system flexible?39.21. Name some of the flexible manufacturing system

software and control functions.39.22. What are the advantages of flexible manufacturing

system technology, compared to conventionalbatch operations?

39.23. Define computer integrated manufacturing.

MULTIPLE CHOICE QUIZ

There are 21 correct answers in the following multiple choice questions (some questions have multiple answers that arecorrect). To attain a perfect score on the quiz, all correct answers must be given. Each correct answer is worth 1 point. Eachomitted answer or wrong answer reduces the score by 1 point, and each additional answer beyond the correct number ofanswers reduces the score by 1 point. Percentage score on the quiz is based on the total number of correct answers.

39.1. Material handling is usually not associated withtransportation between facilities that involves rail,truck, air, or waterway delivery of goods: (a) true or(b) false?

39.2. Fixed routing is associated with which of the fol-lowing types of manufacturing systems (two bestanswers): (a) automated production lines, (b) au-tomated storage systems, (c) cellular manufactur-ing systems, (d) flexible manufacturing systems,(e) job shops, and (f) manual assembly lines?

39.3. Which of the following types of material handlingequipment are typically used in a process type layout(two best answers): (a) conveyors, (b) cranes andhoists, (c) fork lift trucks, and (d) rail-guidedvehicles?

39.4. Batch model production lines are most suited towhich one of the following production situations:(a) job shop, (b) mass production, or (c) mediumproduction?

39.5. Precedenceconstraintsarebestdescribedbywhichoneof the following: (a) launching sequence in a mixedmodel line, (b) limiting value of the sum of elementtimes that can be assigned to a worker or station,(c)orderofworkstationsalongtheline,or(d)sequencein which the work elements must be done?

39.6. Which of the following phrases are most appropri-ate to describe the characteristics of tasks that are

performed at automated workstations (three bestanswers): (a) complex, (b) consists of multiplework elements, (c) involves a single work element,(d) involves straight-line motions, (e) requires sen-sory capability, and (f) simple?

39.7. The transfer line is most closely associated withwhich one of the following types of productionoperations: (a) assembly, (b) automotive chassisfabrication, (c) machining, (d) pressworking, or(e) spot welding?

39.8. A dial indexing machine uses which one of thefollowing types of workpart transfer: (a) asynchro-nous, (b) continuous, (c) parts passed by hand, or(d) synchronous?

39.9. Production flow analysis is a method of identifyingpart families that uses data from which one of thefollowing sources: (a) bill of materials, (b) engi-neering drawings, (c) master schedule, (d) produc-tion schedule, or (e) route sheets?

39.10. Most parts classification and coding systems arebased on which of the following types of partattributes (two best answers): (a) annual produc-tion rate, (b) date of design, (c) design, (d) man-ufacturing, and (e) weight?

39.11. What is the dividing line between a manufacturingcell and a flexible manufacturing system: (a) twomachines, (b) four machines, or (c) six machines?


E1C39 11/11/2009 16:54:58 Page 943

39.12. Amachine capable ofproducingdifferent part stylesin a batch mode of operation qualifies as a flexiblemanufacturing system: (a) true or (b) false?

39.13. The physical layout of a flexible manufacturingsystem is determined principally by which one ofthe following: (a) computer system, (b) materialhandling system, (c) part family, (d) processingequipment, or (e) weight of parts processed?

39.14. Industrial robots can, in general, most easily handlewhich one of the following part types in a flexiblemachining system: (a) heavy parts, (b) metal parts,

(c) nonrotational parts, (d) plastic parts, or (e)rotational parts?

39.15. Flexible manufacturing systems and cells are gen-erally applied in which one of the following areas:(a) high-variety, low-volume production, (b) lowvariety, (c) low volume, (d) mass production,(e) medium-volume, medium-variety production?

39.16. Which one of the following technologies is mostclosely associated with flexible machining systems:(a) lasers, (b) machine vision, (c) manual assemblylines, (d) numerical control, or (e) transfer lines?

PROBLEMS

Manual Assembly Lines

39.1. A manual assembly line is being designed for aproduct with annual demand ¼ 100,000 units. Theline will operate 50 wk/yr, 5 shifts/wk, and 7.5 hr/shift. Work units will be attached to a continuouslymoving conveyor. Work content time ¼ 42.0 min.Assume line efficiency ¼ 0.97, balancing efficiency¼ 0.92, and repositioning time ¼ 6 sec. Determine(a) hourly production rate to meet demand,(b) number of workers required, and (c) the num-ber of workstations required if the estimated man-ning level is 1.4.

39.2. A manual assembly line produces a small appliancewhose work content time ¼ 25.9 min. Desired pro-duction rate ¼ 50 units/hr. Repositioning time ¼ 6sec, line efficiency ¼ 95%, and balancing efficiencyis 93%. How many workers are on the line?

39.3. A single model manual assembly line produces aproduct whose work content time ¼ 47.8 min. Theline has 24 workstations with a manning level ¼1.25. Available shift time per day ¼ 8 hr, butdowntime during the shift reduces actual produc-tion time to 7.6 hr on average. This results in anaverage daily production of 256 units/day. Reposi-tioning time per worker is 8% of cycle time. De-termine (a) line efficiency, (b) balancing efficiency,and (c) repositioning time.

39.4. A final assembly plant for a certain automobilemodel is to have a capacity of 240,000 units annually.The plant will operate 50 wk/yr, 2 shifts/day, 5 days/wk, and 8.0 hr/shift. It will be divided into threedepartments: (1) body shop, (2) paint shop, (3) trim-chassis-final department. The body shop welds thecar bodies using robots, and the paint shop coats thebodies. Both of these departments are highly auto-mated. Trim-chassis-final has no automation. Thereare 15.5 hr of direct labor content on each car in thisdepartment, where cars are moved by a continuous

conveyor. Determine (a) hourly production rate ofthe plant, (b) number of workers and workstationsrequired in trim-chassis-final if no automated sta-tions are used, the average manning level is 2.5,balancing efficiency ¼ 93%, proportion uptime ¼95%, and a repositioning time of 0.15min is allowedfor each worker.

39.5. A product whose total work content time¼ 50 minis to be assembled on a manual production line.The required production rate is 30 units/hr. Fromprevious experience with similar products, it isestimated that the manning level will be close to1.5. Assume that the uptime proportion and linebalancing efficiency are both ¼ 1.0. If 9 sec will belost from the cycle time for repositioning, deter-mine (a) the cycle time and (b) the numbers ofworkers and stations that will be needed on theline.

39.6. A manual assembly line has 17 workstations withone operator per station. Total work content timeto assemble the product ¼ 22.2 min. The produc-tion rate of the line ¼ 36 units/hr. A synchronoustransfer system is used to advance the productsfrom one station to the next, and the transfertime ¼ 6 sec. The workers remain seated alongthe line. Proportion uptime ¼ 0.90. Determine thebalance efficiency.

39.7. A production line with four automatic worksta-tions (the other stations are manual) produces acertain product whose total assembly work contenttime ¼ 55.0 min of direct manual labor. The pro-duction rate on the line is 45 units/hr. Because ofthe automated stations, uptime efficiency ¼ 89%.The manual stations each have one worker. It isknown that 10% of the cycle time is lost due torepositioning. If the balancing efficiency ¼ 0.92 onthe manual stations, find (a) cycle time, (b) number

Problems 943

E1C39 11/11/2009 16:54:59 Page 944

of workers and (c) workstations on the line.(d) What is the average manning level on theline, where the average includes the automaticstations?

39.8. Production rate for a certain assembled productis 47.5 units/hr. The total assembly work contenttime ¼ 32 min of direct manual labor. The lineoperates at 95% uptime. Ten workstations havetwo workers on opposite sides of the line so thatboth sides of the product can be worked on simulta-neously. The remaining stations have one worker.Repositioning time lost by each worker is 0.2 min/cycle. It is known that the number of workers on theline is two more than the number required forperfect balance. Determine (a) number of workers,

(b) number of workstations, (c) the balancing effi-ciency, and (d) average manning level.

39.9. The total work content for a product assembled ona manual production line is 48 min. The work istransported using a continuous overhead conveyorthat operates at a speed of 3 ft/min. There are24 workstations on the line, one-third of whichhave two workers; the remaining stations eachhave one worker. Repositioning time per workeris 9 sec, and uptime efficiency of the line is 95%. (a)What is the maximum possible hourly productionrate if line is assumed to be perfectly balanced? (b)If the actual production rate is only 92% of themaximum possible rate determined in part (a),what is the balance efficiency on the line?

Automated Production Lines

39.10. An automated transfer line has 20 stations andoperates with an ideal cycle time of 1.50 min.Probability of a station failure¼ 0.008 and averagedowntime when a breakdown occurs is 10.0 min.Determine (a) the average production rate and(b) the line efficiency.

39.11. A dial-indexing table has six stations. One station isused for loading and unloading, which is accom-plished by a human worker. The other five performprocessing operations. The longest process takes 25sec and the indexing time ¼ 5 sec. Each station hasa frequency of failure ¼ 0.015. When a failureoccurs it takes an average of 3.0 min to makerepairs and restart. Determine (a) hourly produc-tion rate and (b) line efficiency.

39.12. A seven-station transfer line has been observedover a 40-hour period. The process times at eachstation are as follows: station 1, 0.80 min; station 2,1.10 min; station 3, 1.15 min; station 4, 0.95 min;station 5, 1.06 min; station 6, 0.92 min; and station7, 0.80 min. The transfer time between stations ¼6 sec. The number of downtime occurrences ¼ 110,and hours of downtime ¼ 14.5 hr. Determine(a) the number of parts produced during theweek, (b) the average actual production rate inparts/hr, and (c) the line efficiency. (d) If the

balancing efficiency were computed for this line,what would its value be?

39.13. A 12-station transfer line was designed to operatewith an ideal production rate ¼ 50 parts/hr. How-ever, the line does not achieve this rate, since theline efficiency ¼ 0.60. It costs $75/hr to operate theline, exclusive of materials. The line operates 4000hr/yr. A computer monitoring system has beenproposed that will cost $25,000 (installed) andwill reduce downtime on the line by 25%. If thevalue added per unit produced ¼ $4.00, will thecomputer system pay for itself within 1 year ofoperation? Use expected increase in revenues re-sulting from the computer system as the criterion.Ignore material costs in your calculations.

39.14. An automated transfer line is to be designed.Based on previous experience, the average down-time per occurrence ¼ 5.0 min, and the probabilityof a station failure that leads to a downtime occur-rence p ¼ 0.01. The total work content time ¼9.8 min and is to be divided evenly amongst theworkstations, so that the ideal cycle time for eachstation ¼ 9.8/n. Determine (a) the optimum num-ber of stations on the line n that will maximizeproduction rate, and (b) the production rate andproportion uptime for your answer to part (a).


Documents

Groover fundamentals modern manufacturing 4th txtbk 11