A framework for assessing the use of lean production practices in manufacturing cells

This article was downloaded by: [The University Of Melbourne Libraries]On: 30 September 2013, At: 00:19Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

International Journal of ProductionResearchPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/tprs20

A framework for assessing the useof lean production practices inmanufacturing cellsTarcisio Abreu Saurin a , Giuliano Almeida Marodin a & José LuisDuarte Ribeiro aa Industrial Engineering and Transportation Department, FederalUniversity of Rio Grande do Sul (DEPROT/UFRGS), Av. OsvaldoAranha 99, 5 andar, Porto Alegre, RS, 90.035-190, BrazilPublished online: 28 May 2010.

To cite this article: Tarcisio Abreu Saurin , Giuliano Almeida Marodin & José LuisDuarte Ribeiro (2011) A framework for assessing the use of lean production practices inmanufacturing cells, International Journal of Production Research, 49:11, 3211-3230, DOI:10.1080/00207543.2010.482567

To link to this article: http://dx.doi.org/10.1080/00207543.2010.482567

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the“Content”) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoever orhowsoever caused arising directly or indirectly in connection with, in relation to or arisingout of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &

http://www.tandfonline.com/loi/tprs20

http://www.tandfonline.com/action/showCitFormats?doi=10.1080/00207543.2010.482567

http://dx.doi.org/10.1080/00207543.2010.482567

Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Dow

nloa

ded

by [

The

Uni

vers

ity O

f M

elbo

urne

Lib

rari

es]

at 0

0:19

30

Sept

embe

r 20

13

http://www.tandfonline.com/page/terms-and-conditions

http://www.tandfonline.com/page/terms-and-conditions

International Journal of Production ResearchVol. 49, No. 11, 1 June 2011, 3211–3230

A framework for assessing the use of lean production practices

in manufacturing cells

Tarcisio Abreu Saurin*, Giuliano Almeida Marodinand Jose Luis Duarte Ribeiro

Industrial Engineering and Transportation Department, Federal University ofRio Grande do Sul (DEPROT/UFRGS), Av. Osvaldo Aranha 99, 5 andar,

Porto Alegre, RS, 90.035-190, Brazil

(Received 28 August 2009; final version received 26 March 2010)

This study introduces a framework for assessing the use of lean production (LP)practices in manufacturing cells (MCs). The development of the frameworkincluded four stages: (a) defining LP practices applicable to MC, based on criteriasuch as the inclusion of practices that workers could observe, interact with anduse on a daily basis; (b) defining attributes for each practice, emphasising thedimensions which were typical of their implementation in LP environments; (c)defining a set of evidence and sources of evidence for assessing the existence ofeach attribute – the sources of evidence included direct observations, analysis ofdocuments, interviews and a feedback meeting to validate the assessment resultswith company representatives; (d) drawing up a model of the relationships amongthe LP practices, based on a survey with LP experts. This model supports theidentification of improvement opportunities in MC performance based on theanalysis of their interfaces. A case study of an MC from an automobile partssupplier is presented to illustrate the application of the framework.

Keywords: cellular manufacturing; lean production; performance measurement

1. Introduction

Due to the increasing spread of the lean production (LP) system, the need has arisen todevelop means to investigate the level of maturity of this system in companies. In fact,compared with the efforts made to address ‘how to become leaner’, the statement ‘howlean is the system’ has received less attention (Wan and Chen 2008).

On the one hand, some studies have addressed the issue of lean assessment throughsurveys (White et al. 1999, Shah and Ward 2003). On the other hand, studies have beendevoted to drawing up proposals for assessing the level of LP implementation that couldbe applied individually in each company, which did not aim to obtain statisticalconclusions based on surveys. Examples of such proposals are: the evaluation model forthe Shingo Prize (Shingo Prize 1998), the Lean Enterprise Model (Lean Enterprise Model1998), the standard J4000 – identification and measurement of the best practices forimplementing lean operations (SAE International 1999) and the studies of Karlsson andAhlstrom (1996), Wan and Chen (2008), and Gurumurthy and Kodali (2009).

*Corresponding author. Email: [email protected]

ISSN 0020–7543 print/ISSN 1366–588X online

� 2011 Taylor & Francis

DOI: 10.1080/00207543.2010.482567

http://www.informaworld.com

Dow

nloa

ded

by [

The

Uni

vers

ity O

f M

elbo

urne

Lib

rari

es]

at 0

0:19

30

Sept

embe

r 20

13

The proposals for the Shingo Prize, the Lean Enterprise Model and the SAE standardsoriginated from initiatives by professional associations, or from those in partnership withresearch institutes, who noted the difficulty companies had in assessing progress while LPwas being implemented. Another common feature is that these proposals emphasise theassessment of principles relating to administrative actions and organisational culture,such as, for example, creating an environment conducive to learning.

What Karlsson and Ahlstrom (1996), and Gurumurthy and Kodali (2009) propose issimilar, as they emphasise the collection and analysis of performance indicators that reflectthe adoption of LP principles. Wan and Chen (2008) developed a measure to quantify theleanness level of a manufacturing system based on a benchmark of ideal leanness obtainedfrom historical data. Although this measure could be applied in various parts of amanufacturing system (e.g., a cell or an assembly line), it has not yet been tested in anyreal-world study (Wan and Chen 2008). It is also worth drawing attention to the study byDoolen and Hacker (2005) who developed a lean assessment tool that could be used bothat plant level and as a survey tool to assess LP use in a sample of companies. This toolencompasses 29 LP practices applicable to the company as a whole and it was applied ina survey of manufacturers from the electronics industry in the USA.

With the exception of the study by Doolen and Hacker (2005), these proposals do notaim to assess lean practices, the implementation of which takes place directly and visiblyon the shop-floor (e.g., visual management). In fact, it is well known that the applicationof lean practices does not guarantee the use of lean principles (Spear and Bowen 1999), theassessment of which is more difficult than the assessment of practices. Nevertheless,organisations should emphasise the use of lean practices at the beginning of their leanefforts, since this is a means to introduce lean thinking, even though its integration into theorganisational culture is a very slow process (Mann 2005). Therefore, assessment of leanpractices, rather than lean principles, makes more sense to companies that are in the earlystages of LP implementation.

Moreover, the existing methods for evaluating the implementation of LP do notemphasise the analysis of the interfaces between lean practices, even though several studieshave stressed that LP must be fully integrated under a systemic view (Liker 2004). In fact,such evaluation is difficult since few models explicitly present the dependence relationshipamong LP practices. Relationships are often over-simplified, as occurs in the well-knownHouse of Toyota Production System (Liker 2004).

It is also worth noting that the existing methods were mostly designed to assess thelevel of LP implementation in the plant as a whole rather than in specific units ofthe manufacturing system, such as cells, job shops or assembly lines. This is an importantgap, since the level of LP implementation can vary within a plant, and the characteristicsof LP to be evaluated may take on a different nature depending on the process type(Lane 2007).

In this context, this study investigates the assessment of lean practices in manufactur-ing cells (MCs), which are widely known as a strategy for shortening lead times, improvingproduct quality, and increasing flexibility (Sheridan 1990). Since these benefits areprioritised in LP environments, MCs are often used in this context (Hyer and Wemmerlov2002). Moreover, the implementation of LP practices might be easier in MCs due to thereduced number of workers and pieces of equipment in comparison with large assemblylines and functional departments.

The main objective of this article is to introduce a framework for assessing the useof LP practices in MCs. The purposes of this framework are twofold: (a) to point out

3212 T.A. Saurin et al.

Dow

nloa

ded

by [

The

Uni

vers

ity O

f M

elbo

urne

Lib

rari

es]

at 0

0:19

30

Sept

embe

r 20

13

strengths and weaknesses in MCs that are intended to be lean; and (b) to support the

identification of improvement opportunities, based on the analysis of the interfaces among

LP practices. An assumption of this study is that the investigation into the relationships

among LP practices tends to be easier in the microenvironment of MC.As an element of the proposed framework, this article also suggests a set of

LP practices that might be used in MCs. A case study is presented to illustrate the

application of the framework in a real MC.It is worth noting that the framework is not intended to provide a full lean assessment

of MCs, since two important dimensions of leanness are neglected: the use of lean

principles and the output measured using certain metrics of the MC. Although a positive

correlation might be expected among the use of lean principles, the use of lean practices,

and the output, this study focuses on the use of practices and does not provide empirical

evidence for either supporting or refuting the correlation mentioned.

2. Attributes of MC

The concepts of cellular manufacturing presented in the literature are often far from

precise (Hyer and Brown 1999). Thus, in this article we set out to establish a set of

attributes of MCs, which in turn is closely based on the studies of Hyer and Brown (1999),

and Hyer and Wemmerlov (2002).

2.1 Group technology attributes

These attributes are concerned with two key features of MCs: (a) operations should be

different and sequential; and (b) operations should be dedicated to a product family (Hyer

and Wemmerlov 2002). It is worth stressing that an MC might consist of pieces of identical

equipment, provided they are used to undertake different and sequential operations

(Shafer et al. 1995). Feature (b) implies that MCs must adopt the group technology (GT)

concept, the main objectives of which are to reduce complexity and to obtain economies of

scale by grouping similar products in the same layout (Burbidge 1996).

2.2 Connections of time, space and information

According to Hyer and Brown (1999), an MC is characterised by the creation of a

workflow in which the tasks required and workers are closely connected in terms of time,

space and information. Connections in time mean that transfer and waiting times between

all cell operations should be minimised. Connections in space imply that all operations

should be in physical proximity to each other, making it easier to move materials, to

exchange information, and to solve problems. Connections in information imply that

information is accurate, complete, and easily accessible (Hyer and Brown 1999). Since the

concept of ‘connections’ is abstract and therefore difficult to directly observe and assess,

Hyer and Brown (1999) identified a set of practices that could ensure there were

connections – many of them are typical LP practices.

International Journal of Production Research 3213

Dow

nloa

ded

by [

The

Uni

vers

ity O

f M

elbo

urne

Lib

rari

es]

at 0

0:19

30

Sept

embe

r 20

13

2.3 Organisational attribute

This attribute recognises an MC as an administrative unit that requires a dedicatedinfrastructure, including its own performance targets, workers and supervisors. However,a dedicated and physically clustered group of resources is not considered an MC if it is notmanaged and monitored separately from other resources (Hyer and Wemmerlov 2002).In fact, a key cellular manufacturing principle is to divide the manufacturing systeminto semi-autonomous subsystems, so each might respond faster to client demands(Satoglu et al. 2006).

3. Research method

3.1 Overview of the research method

The method used to evaluate the implementation of LP practices in MCs canbe understood as an audit, which is a systematic, independent and documented processto obtain evidence and to evaluate it objectively in order to determine the extent to whichaudit criteria are met (Chiesa et al. 1996). According to Chiesa et al. (1996), auditingallows the gap between desired and actual performance to be identified, as well as enablinginformation to be provided that can be used to improve performance. In severaldimensions of performance (e.g., quality), audits are widely used to identify the presence ofbest practices. For this research, best practices include a set of practices that areconsistently listed in the literature as effective in operationalising LP principles in MCs.

The development of the framework included four stages: (a) defining LP practicesapplicable to MCs; (b) defining attributes for each practice; (c) defining a set of evidenceand sources of evidence for assessing the existence of each attribute; and (d) drawing up amodel of the relationships among the LP practices, based on a survey with LP experts.This model supports identification of improvement opportunities in MC performancebased on the analysis of their interfaces.

After drafting the framework, it was tested and improved by applying it to one cell(referred to as MC1) of a manufacturer of automobile engine components. Thismanufacturer has adopted LP as a corporate policy and the research team had easyaccess to the data necessary for conducting the study, since one of the researchers hadpreviously undertaken both research and consultancy projects on LP in that organisation.The case study included interviews with company representatives after the framework wasapplied in order to obtain their feedback on its perceived strengths and weaknesses.

3.2 Definition of LP practices applicable to MCs and their attributes

Three criteria were adopted to define the LP practices that would be assessed in an MC:(a) the contribution of the practice to the implementation of the MC attributes, based onthe literature review; (b) the inclusion of practices that workers could observe, interactwith and use on a daily basis; and (c) the exclusion of practices that were not implementedwithin the physical boundaries of the cell. Thus, some typical LP practices were notincluded, such as simultaneous engineering. Of course, from a system’s perspective, there isthe need to recognise that the omitted practices also have an impact on MC performance.

As a general guideline for defining the practices, it was established that they shouldtake a systemic view of MCs, involving both their technical and human elements. Indeed,this was an appropriate choice since most cell designs have a technical focus, and thus give


Dow

nloa

ded

by [

The

Uni

vers

ity O

f M

elbo

urne

Lib

rari

es]

at 0

0:19

30

Sept

embe

r 20

13

limited attention to the significant human dimensions (Hyer et al. 1999). Thus, it wasdefined that the practices should be grouped into three of the subsystems that form amanufacturing system (Figures 1–3): human resources, production planning andcontrol (PPC), and process technology. These three subsystems were identified based ona list of decision-making factors in manufacturing systems proposed by Miltenburg (2005),which was adapted to the context of MCs. The adopted subsystems are similar to thesubsystems of a socio-technical system, based on the classification put forward byHendrick and Kleiner (2001): (a) a technical subsystem, which equates to the processtechnology subsystem; (b) a personnel subsystem, which equates to the human resourcessubsystem; and (c) a work design subsystem, which equates to the PPC subsystem.Hendrick and Kleiner (2001) add that the external environment (formed by socio-economic, educational, political, legal, and cultural factors) is another key component of asocio-technical system, since it strongly influences the other three subsystems. From theMC perspective, the external environment encompasses both the support areas whoseactivities were not included in the list of selected practices (e.g., the purchasingdepartment), and the factors mentioned by Hendrick and Kleiner (2001). As a

Human resources Practices Attributes Main sources

1. Teamwork and leadership (TWL)

Team leader supports workers in continuous improvement activities, such as problem solving and implementation of improvements. Team leader substitutes missing workers. Performance assessment of workers is made on a team basis, rather than on an individual basis.

Liker (2004); Mann (2005)

2. Continuous improvement (CI)

Workers are trained in problem solving methods, including root cause analysis.Workers are involved in continuous improvement initiatives, whether formal or informal ones. Continuous improvement groups are coordinated either by workers or team leaders.

Liker (2004); Mann (2005)

3. Multi-functionality and cross-training(MCT)

All workers are able to carry out all cell operations (i.e. cross-training is fully implemented). There is a skills matrix that documents every worker’s skills. Job rotation among cell workstations is undertaken on a daily basis.

Monden (1998); Miltenburg (2001)

4. Workers’ autonomy (WAU)

Workers have autonomy both to identify and to control process and product variability. Workers have autonomy to stop production if abnormalities occur. There are visual devices for calling the team leader or support areas, such as maintenance.

Monden (1998); Liker (2004)

5. Standardised work (STW)

There are documented work standards. Work standards are visible to the team leader. Standards include information on takt time, cycle times, manual and automatic time, production sequence, standard inventories, and cell layout. Standards are updated on a regular basis. There are audits to check compliance with work standards on a regular basis.

Spear and Bowen (1999); Miltenburg (2001)

6. Workplace housekeeping (WHK)

The cell is clean and equipped with only the necessary objects. Every object has a standard place, which is easily identified by visual devices.There is a 5S program, which is audited on a regular basis. Results of 5S audits are posted in the cell.

Monden (1998); Liker (2004)

Figure 1. Attributes of LP practices in the human resources subsystem.


Dow

nloa

ded

by [

The

Uni

vers

ity O

f M

elbo

urne

Lib

rari

es]

at 0

0:19

30

Sept

embe

r 20

13

hypothetical example of the impact of these factors, can be mentioned the possible

shortage of qualified labour in certain regions (this would be a factor associated with thesocio-economic and educational external environment), which in turn could make itdifficult to implement some LP practices, such as continuous improvement.

Based on the criteria and guidelines proposed, 18 LP practices applicable to MCs were

defined. Indeed, many of these practices were not directly selected and extracted from theliterature, since they had to be restated in order to fit the boundary conditions that were

put forward. For instance, the LP practice referred to as ‘autonomation’ was separatedinto two practices: one of them emphasising its technical dimension, as a practice labelled‘equipment autonomation’, and the other emphasising its human dimension, as an

attribute of the practice ‘workers autonomy’.Once the practices were defined, a set of attributes was established for each of them,

based on classical studies of LP. Similarly to the definition of the practices, theassumptions underlying the selection of their attributes need to be stressed:

(1) The attributes emphasise the dimensions of each practice which are typical of theirimplementation in LP environments. For instance, the existence of interpersonalskills which enable people to work well in teams is a desirable characteristic of

teamwork that was not included, since it was assumed interpersonal skills aredesirable whether or not the cell is intended to be lean;

Production planning and control

Practices Attributes Main sources

7. Pull production (PULL)

All inventories (raw materials, work-in-process, and end products) have visually defined maximum caps. There are visual devices informing both production sequencing and materials loading sequences. There are standard routes for loading raw materials and removing end products, including standard picking times. The above attributes exist for all components, whether manufactured in the plant or purchased from external suppliers.

Smalley (2004); Liker (2004)

8. Smoothed production (SPR)

All product models are produced every day. Consumption of raw materials from the preceding processes occurs at constant intervals and volumes.

Womack and Jones (1998); Liker (2004)

9. Quick setups (QST)

There are no setups among different models. If there is setup, its tasks are standardised and separated into internal and external tasks

Shingo (1989)

10. Total Productive Maintenance (TPM)

Workers carry out routine maintenance on all equipment (e.g. cleanliness, lubrication and small repairs) following standardised procedures. There is either preventive or predictive maintenance of all equipment.

Black (2007)

11. Lean performance metrics (LME)

Cell performance is assessed based on metrics linked to lean production principles, such as lead time, rework and scrap rates, standard inventory versus actual inventory, overall equipment effectiveness (OEE).

Rother and Harris (2001); Maskell and Baggaley (2004)

12. Visual management of production control (VPC)

There is a production control board visible to all cell workers, showing production schedule either on an hourly or shift basis. The following information is presented on the board: planned; undertaken; difference pending; reasons for failing to comply with schedule; corrective actions.

Maskell and Baggaley (2004); Mann (2005)

Figure 2. Attributes of LP practices in the production planning and control subsystem.


Dow

nloa

ded

by [

The

Uni

vers

ity O

f M

elbo

urne

Lib

rari

es]

at 0

0:19

30

Sept

embe

r 20

13

(2) The behavioural dimensions of each practice were left aside (e.g., do workers eitheroverproduce or keep the equipment idle when there is no demand from upstreamprocesses?), since assessing these would make more sense in an evaluation of theextent to which the organisational culture is lean;

(3) The attributes emphasise the existence of practices rather than the result of thosepractices. For instance, assessment is made in terms of whether or not systematicactivities dedicated to continuous improvement are in place, rather than in terms oftangible outputs of continuous improvement activities, such as improved quality;

(4) The attributes emphasise only the operational aspects of each practice that can beassessed fairly quickly and using sources of evidence that are usually easilyavailable (e.g., observations). Thus, even though comprehensive labels such asTotal Productive Maintenance (TPM) and pull production are used, the attributesdo not encompass the full complexity of each practice.

3.3 Evidence and sources of evidence for assessing the use of the LP practices

3.3.1 Characterisation of the company and the MC under study

Before assessing the LP practices, it is important to understand the organisational contextin which the MC is placed as well as to identify the extent to which the attributes of theMC are in place, based on the concepts presented in Section 2. This analysis is importantto validate whether the framework is being used for the purpose for which it was primarilydesigned (i.e., the assessment of lean practices in MCs), rather than to assess LP in other

Process technology

Practices Attributes Main sources

13. Visual management of quality control (VQC)

There are quality control boards, which are visible to all cell workers. The boards display quality related metrics, root causes of defects, and respective action plans.

Black (2007); Maskell and Baggaley (2004)

14. Equipment autonomation (EQA)

Machinery carries out value adding operations without either workers monitoring them or manual intervention. All pieces of equipment have devices either for preventing or detecting abnormalities. These devices stop production or provide warning of abnormalities.

Shingo (1989); Monden (1998)

15. One-piece-flow (ONE)

Single pieces of material are produced and moved between operations. There is no piece of material waiting between adjacent workstations.

Rother and Harris (2001); Liker (2004)

16. Visibility and information exchange (VIS)

All workers can easily see their cell counterparts, equipment and materials. All workers can talk with each other in a normal tone of voice.

Womack and Jones (1998); Hyer and Wemmerlov (2002)

17. Layout size and shape (LSS)

All workers can exchange materials without walking more than 1 m (this distance was arbitrarily established). Cell design allows changing the amount of workers and production capacity.

Monden (1998); Hyer and Wemmerlov (2002)

18. Organisation by the dominant flow (ODF)

All products pass through the same processes in the same sequence.

Miltenburg (2001); Rother and Harris (2001)

Figure 3. Attributes of LP practices in the process technology subsystem.


Dow

nloa

ded

by [

The

Uni

vers

ity O

f M

elbo

urne

Lib

rari

es]

at 0

0:19

30

Sept

embe

r 20

13

types of processes. The lack of evidence of MC attributes provides a preliminary indicationthat the use of LP might be compromised, especially for those attributes that are clearlycell specific, such as layout size and shape.

A form was designed to guide data collection during this stage. It consisted of 30questions spread over four sections: (a) questions about the plant, such as its size,products, customers and current and previous LP initiatives; (b) questions about the cell,such as the number of operators, the range of equipment and products, methods ofsupervision, performance metrics, and resource sharing with other cells or departments;(c) questions designed to obtain evidence regarding the time, space and informationconnections (e.g., are the operators close to each other physically? Can they talk with eachother in a normal tone of voice?) – these questions, unlike the others on this form, werepresented as a check-list requiring either ‘yes’ or ‘no’ answers; and (d) development of theproducts versus the process matrix as well as the calculation of the grouping efficiencyindex, to assess the group technology (GT) attribute. Grouping efficiency was calculatedbased on the method proposed by Chandrasekharan and Rajagopalan (1989), whichconsiders both intracellular operations (manufacturing steps that occur inside the cell andthat should be maximised) and extra cellular operations (manufacturing steps that occurout of the cells and that should be minimised).

3.3.2 Assessment of the use of LP practices

Assessing the use of LP practices was supported by a set of data collection forms:

(1) Form A has 22 questions related to practices that can be visually identified(Figure 4) and is a guide for direct observations of the MC. For some questions ofthis form, the auditor was requested to carry out other tasks than just that ofobserving, such as drawing a sketch of the cell layout and drawing a simplifiedvalue stream map for the product family processed in the MC.

(2) Form B (Figure 5) has 13 questions and is a guide for interviewing operators. It canbe applied either on an individual basis or to groups.

(3) Form C has 29 questions (Figure 6) and it is a guide for interviewing the MCteam leader or, if this role does not exist, the operators’ immediate superior.

Yes No(1) Are there documented work standard forms? If the answer is yes, point out what information is included in the forms: ( ) work sequence; ( ) cycle time; ( ) takt time; ( ) standard WIP; ( ) cell layout (2) Are the work standard forms easily visible for team leaders and auditors of the compliance with standards? * if the forms are only available in computers, then mark ″no″(3) Are there visual devices (e.g. andons) that can be used either by workers or team leaders to request help from support areas (e.g. maintenance, quality)? (4) Does the inventory of finished products have caps, established either by physical or visual barriers? * if the answer was yes, please point out if: ( ) it is a supermarket type of inventory, since there is replenishment of products identical to the ones that were consumed by the subsequent process ( ) it is a FIFO type of inventory, since products are consumed by the subsequent process on a FIFO basis and consumption does not necessarily mean that an identical product should be replenished (5) Is there a production control board visible to all cell workers?* If the answer was yes, what is the time span adopted in production control (i.e. filling out the production control board)? ( ) hourly ( ) daily ( ) shift ( ) other, please specify in this case:____________

Figure 4. Excerpt from Form A, used to support cell observation.


Dow

nloa

ded

by [

The

Uni

vers

ity O

f M

elbo

urne

Lib

rari

es]

at 0

0:19

30

Sept

embe

r 20

13

Many questions on this form are the same as those included on Form B, thusproviding an opportunity to compare the answers by workers with those of theteam leaders.

The proposed assessment forms are fully in line with the LP principle of investigatingproblems in the place where they happen, looking at the process and talking with thepeople involved. As they are couched in lean jargon, FormsA, B and C force the auditor togo to the gemba, which can be roughly translated from the Japanese as the place where theaction is happening (Mann 2005).

The assessment of LP practices was subsequently carried out based on the informationcollected via Forms A, B and C. To facilitate linking the forms to the LP practices so thatthe latter could be assessed, a matrix was compiled, as is partially illustrated in Figure 7.The results of the assessment were classified into one of three situations: (a) no compliance,if the cell did not comply with any of the attributes; (b) partial compliance, if the cellcomplied with one or more attributes, but not all of them; and (c) full compliance, if thecell complied with all the attributes established.

Thus, compliance with an attribute means that the cell adheres to all requirementspresented in each sentence of the column ‘attributes’ in Figures 1–3. If compliance withpart of any sentence is lacking, the whole practice should be classified as partiallycomplying. For instance, one of the attributes presented in Figure 3 is that there should bequality control boards visible to all cell workers. If the cell had boards, but badlypositioned ones, the associated LP practice would be classified as partially complying, evenif the remaining attributes were in place.

3.4 Development of a relationship model of the interfaces among LP practices

The model was developed based on a questionnaire administered to seven LP experts. All ofthem have more than five years’ experience of teaching LP and acting as LP consultants.One of them is also a coordinator of continuous improvement in a large company whichmanufactures automotive components and has adopted LP as a corporate policy.

Yes No(1) Does the team leader (or supervisor) provide support (e.g. substituting missing workers, solving problems) when requested? (2) Is the performance assessment conducted either ?sisabmaetarosisablaudividninano(3) Are you authorized and encouraged to stop producing when any abnormality is detected? If you answered yes, point out typical abnormalities that demand production stoppages:_________ (4) Are you trained to carry out all operations in this cell? * if the answer was no, please point out in which operations you are trained:____________ (5) Are you in charge of both routine and basic preventive maintenance tasks (e.g. cleaning and visual inspection) in the equipment? * If the answer was yes, please specify if these are standards for these tasks.

Figure 5. Excerpt from Form B, used to support interviews with cell workers.

Yes No(1) Do you both substitute missing workers a ?detseuqernehwsrekrowplehdn(2) Is there any preventive maintenance program? * If the answer is yes, point out which cell equipment are encompassed by the maintenance program:__________ (3) Are workers requested to identify and control defective production? * If the answer is yes, point out what are the major causes of defects:_____________________________ (4) Are there workers trained to carry out all operations in this cell?(5) Do you have any means of formal control over the skills of each worker in this cell, such as a skills matrix?

Figure 6. Excerpt from Form C, used to support interviews with the team leader.


Dow

nloa

ded

by [

The

Uni

vers

ity O

f M

elbo

urne

Lib

rari

es]

at 0

0:19

30

Sept

embe

r 20

13

The experts filled out a questionnaire which had the following question: in the context ofMCs, to what extent does practice X support the implementation of practice Y? Since therewere 18 practices (see Figures 1–3), n(n� 1) or 306 answers were required.

The experts were asked to indicate how strong each interface was, based on a fourpoint discrete scale: 0 (no support); 1 (weak support); 2 (moderate support); 3(strong support). Together with the questionnaire, all experts received a list of theattributes of LP practices, and their definitions, adopted in this study. The completion ofthe questionnaire was a fairly time-consuming and mentally demanding task (about 90min-utes if 15 seconds/question). Due to this fact, it was not completed all at once by theexperts; they answered the questions over a few days. The nature of this task was the mainreason why these same experts were not asked to validate the list of LP practices andattributes. Indeed, the trade-off on how best to use the experts’ time was decided in favourof the relationship model since, unlike for the list of LP practices and attributes, theassignment of scores could not be supported by the literature review.

Based on the survey results, a model was drawn up which considers relations that hadinterface strength averages equal to or higher than 2.0. This cut-off point was establishedin order both to emphasise the strongest interfaces and to make it easier to understand themodel, since it would be too confusing if all interfaces were shown. Although analternative model considering just averages higher than 2.0 was also drawn up, it was notadopted since some important relationships would be missing, particularly those related tothe human resources subsystem. It is worth stressing that the final model was obtainedonly after several iterations and a meeting with the experts who answered the survey.In this meeting the experts explained the reasons for their answers, even though no attemptwas made to achieve a consensus and their original scoring was respected.

3.5 Application of the framework to the industrial case

3.5.1 Preparatory stage

This stage included three activities: (a) introducing the assessment procedures to thecompany’s representatives, including cell workers, aiming at obtaining their collaborationduring data gathering and analysis; (b) choosing the cell to be assessed; and (c) negotiatingthe assessment schedule. For the case study, MC1 was chosen because its processes were oflow complexity, thus making the first experience of using the assessment tools easier.Moreover, since MC1 is similar to many other cells in the plant, some of the conclusionsof the assessment could have an impact beyond the specific cell being analysed.

3.5.2 Data collection

Characterising both the company and MC1 was based on interviews with the plantdirector, the supervisor and the coordinator of LP implementation, with Form I beingused as a guide for data collection. On average, each interview took about 45minutes.

Practices Sources of evidence (forms and questions) Teamwork and leadership Form B (questions 1,2) and Form C (question 1) Workers autonomy Form A (question 3), Form B (question 3) and Form C (question 3)

Figure 7. Example of linking data collection forms to LP practices so that the latter could beassessed.


Dow

nloa

ded

by [

The

Uni

vers

ity O

f M

elbo

urne

Lib

rari

es]

at 0

0:19

30

Sept

embe

r 20

13

The interviews used for completion of Forms B and C were undertaken away from thecell, in a place where the interviewees could have privacy to express their opinions. Whilethe interviews with three cell workers took about 30minutes (they were interviewed as agroup), the interview with the supervisor took about 45minutes.

In the case study, observations of MC1 in operation (supported by FormA) tookabout 90minutes and they were made jointly with the supervisor, who also answered thequestions on Form C.

3.5.3 Feedback meeting

This stage included a meeting with the company’s representatives to discuss the results ofthe assessment and to identify opportunities for improving performance. The meeting wascoordinated by a member of the research team and was attended by all stakeholdersinvolved in the assessment. The practices that did not comply with the attributes, bothpartially or fully, were analysed and discussed from two perspectives: (a) checking previousexperiences with the practices, since this could aid in the understanding of why currentattempts failed; and (b) considering the interfaces with other practices.

4. Results

4.1 Relationship model

Table 1 presents the average intensity of each of the relationships, as defined by the leanexperts, emphasising those that are either equal to or higher than 2.0. Table 2 presents thenumber of relationships of each practice, only considering those with an average intensityequal to or higher than 2.0. Based on the data presented in Table 2, practices wereclassified into four categories: end practices, intermediate practices, basic practices, andstand-alone practices. End practices are those that depend on other practices at least twiceas frequently as they support other practices. Basic practices are those that support otherpractices at least twice as frequently as they depend on other practices. Intermediatepractices depend as much on other practices as other practices depend on them. The stand-alone practices did not have strong relationships with any other practices. Thisclassification does not imply that practices from some categories are intrinsically moreimportant than practices from other categories. Importance should be assessed on anindividual basis for each cell, since it depends on contextual factors such as the extentto which each practice contributes to achieving the performance targets of the cell(e.g., workers’ satisfaction and on-time delivery).

Figure 8 shows a graphical illustration of the relationships presented in Table 1. Thetwo practices that do not appear in Figure 8 are equipment autonomation (EQA) and leanperformance metrics (LME), which were classified as stand-alone practices. It is worthnoting that three end practices (PULL, CI and ONE) might be directly associated withthree LP principles established by Womack and Jones (1998), referred to as pull,perfection, and flow. The two other principles established by Womack and Jones (1998),referred to as specifying value from the standpoint of the end customer and wasteelimination along the value stream, are not explicit in the model. On the one hand, themodel does not encompass product design, which is the primary opportunity to identifyproduct features that add value from the perspective of the end-customer. On the other


Dow

nloa

ded

by [

The

Uni

vers

ity O

f M

elbo

urne

Lib

rari

es]

at 0

0:19

30

Sept

embe

r 20

13

Table

1.Averageintensity

oftherelationshipsamongLPpractices.

Practices

asindependentvariables

TWL

CI

MCT

WAU

STW

WHK

PULL

SPR

QST

TPM

LME

VPC

VQC

EQA

ONE

VIS

LSS

ODF

Practicesasdependentvariables

TWL

X1.3

22

1.4

0.7

10.6

0.9

11.6

21.7

0.7

0.4

1.7

1.4

0.4

CI

2.9

X2

22.6

1.6

1.3

0.9

1.1

1.6

1.9

2.6

2.3

1.6

12.1

0.7

0.9

MCT

20.9

X1.4

2.7

1.3

0.6

0.9

0.7

1.1

10.3

0.3

1.9

0.3

11.3

0.7

WAU

1.6

1.4

2X

21.1

1.4

0.7

0.6

1.4

1.1

1.4

1.4

1.7

0.4

21

0.7

STW

1.1

2.1

11.3

X2

1.1

1.1

1.3

0.9

1.1

1.4

1.3

1.1

0.3

10.7

1WHK

1.9

21.4

1.3

2.7

X1.1

0.9

1.3

20.9

0.9

1.3

0.3

0.7

1.1

1.7

1.4

PULL

1.7

1.4

1.1

12

1.3

X2.7

2.9

2.3

1.3

2.1

2.1

1.3

2.9

12

2SPR

0.9

1.1

1.1

0.7

1.4

12.1

X2.9

21

1.4

1.6

12.6

0.6

0.9

1.7

QST

1.9

2.3

0.7

0.9

1.7

1.9

1.6

1.4

X1.1

1.1

0.6

0.4

0.9

0.6

0.7

0.6

0.6

TPM

21.4

1.4

1.7

2.1

2.3

11.7

1.3

X1

1.4

1.6

0.4

0.4

0.7

0.4

0.3

LME

11

0.7

0.7

1.9

0.9

10.9

11.3

X1.1

1.3

0.3

0.7

1.1

0.1

0.3

VPC

1.6

1.1

0.6

0.9

1.7

1.3

1.6

1.1

0.4

0.6

1.1

X1.4

0.9

1.7

2.3

1.4

0.9

VQC

1.3

1.3

0.4

11.9

1.1

1.3

0.7

0.4

0.6

1.3

1.9

X1.1

1.3

2.3

1.3

0.6

EQA

0.9

1.7

0.6

0.4

0.9

0.4

10.7

0.6

1.4

0.1

0.1

0.1

X0.4

0.1

0.1

0.6

ONE

0.9

1.6

1.6

1.6

2.3

0.9

2.3

1.7

2.9

20.9

1.3

0.9

1.3

X1.3

2.3

2.4

VIS

1.4

1.1

0.9

11.1

1.9

1.4

0.7

0.6

0.7

1.4

1.4

1.6

0.9

1.3

X2.3

1.7

LSS

0.3

1.4

1.1

0.7

11.4

1.7

1.1

1.6

0.7

0.4

0.6

0.6

0.9

1.7

1.3

X2.6

ODF

0.3

0.9

0.1

0.3

0.3

0.9

1.1

1.9

10.1

0.7

0.9

0.3

0.1

1.3

0.4

1.7

X


Dow

nloa

ded

by [

The

Uni

vers

ity O

f M

elbo

urne

Lib

rari

es]

at 0

0:19

30

Sept

embe

r 20

13

Figure 8. Relationship model of the dependence among LP practices.

Table 2. Number of relationships with an average intensity equal to or higher than 2.0.

Practice Depends on . . . Supports . . . Sum of relations Classification

ODF 0 3 3 BasicSTW 2 7 9 BasicVIS 1 4 5 BasicQST 1 3 4 BasicVPC 1 3 4 BasicLSS 1 3 4 BasicVQC 1 2 3 BasicMCT 2 3 5 IntermediateTPM 3 4 7 IntermediateTWL 3 3 6 IntermediateWAU 3 2 5 IntermediateWHK 3 2 5 IntermediateONE 6 3 9 End practiceCI 7 3 10 End practiceSPR 4 1 5 End practicePULL 9 2 11 End practice


Dow

nloa

ded

by [

The

Uni

vers

ity O

f M

elbo

urne

Lib

rari

es]

at 0

0:19

30

Sept

embe

r 20

13

hand, the model emphasises the use of lean practices at individual stages of a value stream(i.e., individual cells), rather than along multiple stages of the value stream.

Furthermore, the emphasis of the model on work standardisation (STW; sevenother practices heavily depend on it) might be viewed as an indirect consideration of thetwo principles related to value. Indeed, a core step of work standardisation methodsfrom the lean perspective is to identify and eliminate non-value adding activities(Rother and Harris 2001).

The model also indicates that the lean experts considered some practices from abroader perspective than the list of attributes. For instance, the assumption thatcontinuous improvement (CI) strongly depends on seven other practices seems to beinconsistent with its attributes which emphasise operational aspects that could beimplemented in isolation (e.g., training on problem solving). However, the interpretationmade by the lean experts was that these aspects would not be enough to sustain continuousimprovement in the long-term, without the support of other practices. For instance, multi-skilling and cross-training make operators both more capable of identifying opportunitiesfor improvement and committed to implementing them.

However, the operational and short-term dimension of continuous improvement wasalso encompassed by the model, since it stresses that CI supports quick setups (QST) andworkplace housekeeping (WHK). In fact, kaizen events may have setup reduction as theirfocus and housekeeping is usually dealt with, whatever the focus of the kaizen event is.

The model also has two important synergistic relationships among the practices PULL,ONE, and SPR. In fact, while PULL contributes to ONE by establishing work-in-processcaps and gradually reducing inventories, ONE is the ideal stage for PULL. As to the othersynergy, SPR implies the maintenance of a stable pattern of mix and volume, ideallychanging models at short intervals and using small batches. The impact of thischaracteristic is that demand should be smoothed throughout upstream processes,which should be able to supply materials for the mixed demand required by downstreamprocesses – this is a characteristic of PULL. In turn, PULL calls for smoothing, since if thedemand far exceeds the planned pattern of mix and volume, there will be peaks andtroughs of demand detrimental to just-in-time replenishment (Smalley 2004).

The model has three other synergistic relationships that are fairly straightforward.While WHK makes it easier to find maintenance tools in a clean and organised workplace,TPM tackles wastes and residues that compromise housekeeping (e.g., oil leakages).Similarly, while WHK supports the detection of deviations from standards, the right placeof every object is determined by standards (STW). The synergy between CI and STW iswell-known since the standards are the basis on which improvements are built, while animprovement can only be sustained if it is standardised (Liker 2004).

Figure 8 also indicates that the model has one loop related to the human subsystem,which involves the practices of workers’ autonomy (WAU), teamwork and leadership(TWL), and multi-functionality and cross-training (MCT). Indeed, problem solvingsupported by the team leader (an attribute of TWL) depends on the existence of visualdevices to call the leader (an attribute of WAU). Moreover, performance assessment on ateam basis (an attribute of TWL) makes more sense if workers are encouraged to requesthelp to solve problems. In turn, granting autonomy in order to detect abnormalities(an attribute of WAU) requires capable workers, which may be obtained by having amulti-functional and cross-trained workforce.

The human loop has a synergy between TWL and MCT. On the one hand, TWLsupports MCT since, with team-based performance assessment, workers are encouraged to


Dow

nloa

ded

by [

The

Uni

vers

ity O

f M

elbo

urne

Lib

rari

es]

at 0

0:19

30

Sept

embe

r 20

13

help each other and, as a result, learn about the tasks conducted by their colleagues.On the other hand, multi-functional and cross-trained workers are more able to help eachother and, thus, team-based assessment makes more sense.

4.2 Case study

4.2.1 Characterisation of the company and the MC under investigation

The company examined has 150 workers and manufactures electrical engine componentsfor the automotive and household appliance industries. The transnational group that tookover the company in 2002 made the decision that LP should be adopted as a corporatepolicy, and this was implemented in all plants around the world. LP implementation wasmade the responsibility of a manager whose job title was Lean Coordinator. Moreover,LP implementation gained momentum when the company was involved in a supplierdevelopment programme led by an automotive manufacturer that was one of its maincustomers.

Cellular manufacturing was introduced in 2003 as part of the first initiatives to applyLP principles, and was aimed at shortening lead times. Although the sequence of processsteps may change from cell to cell, all cells in this plant are similar to MC1 which producesbrushes and was designed to be staffed with from one to three workers.

Each worker carries out all operations in identical and full circuits, which characterisesthe rabbit-chase job assignment (Hyer and Wemmerlov 2002). Figure 9 shows MC1 withtwo workers. Production starts by welding two components that are stored in cradles.Then, other components are assembled and another welding operation is undertaken tojoin the previous mounts with a block, which is an internal component manufactured inthe stamping department. Cycle times are less than 20 seconds and workers hold theproducts in their hands during all operations, including welding.

With regard to GT, the grouping efficiency of this cell is 100%, which means that all23 models have identical flows. As to space, the cell occupies about 8m2 and it has smallpieces of equipment. Connections of space, as well as connections of time and information,are strengthened by one-piece-flow and the existence of only three operations. The rabbit-chase strategy also strengthens time and information connections, since any substantialvariation in individual cycle times immediately impacts on other workers’ performance. Bycontrast, MC1 has shortcomings regarding the organisational attribute, such as workersnot being fully dedicated to the cell (they rotate among the remaining cells within thedepartment).

4.2.2 Analysis of LP practices in MC1

In MC1, seven lean practices are fully used (MCT, WHK, VPC, ONE, VIS, LSS, andOFD), three are partially used (WAU, STW, and LME), and eight are not used (TWL, CI,PULL, SPR, TPM, VQC, QST, and EQA). To some extent, this situation can beattributed to the excessive dependence on the lean coordinator and to the fact that LPimplementation was driven by external impositions (i.e., corporate policy and customerdemands), rather than by top management taking a real interest. Figure 10 exemplifieshow the results were compiled and presented to the company’s representatives, whileFigure 11 shows the relationship model after taking the results of the assessment intoconsideration.


Dow

nloa

ded

by [

The

Uni

vers

ity O

f M

elbo

urne

Lib

rari

es]

at 0

0:19

30

Sept

embe

r 20

13

As shown in Figure 11, practices that are dependent on each other do not necessarilyshare the same level of compliance. This might happen since there can be other variablesimpacting the level of compliance, variables which go beyond the dependencies among thepractices and which are not captured by the model. In fact, the relationship modelpresented in Figure 8 should not be considered complete and comprehensive.

In some cases, practices in full compliance followed by no complying practices indicatethat it will be fairly easy to improve performance, since important requirements are inplace. For instance, this type of relationship did exist between visibility and informationexchange (VIS, full compliance) and visual management of quality control (VQC, nocompliance). Indeed, since all cell workers have full visibility of the whole cell (an attributeof VIS), quality control boards could probably easily be positioned in a place visible to thewhole cell team (an attribute of VQC). However, quality control boards did not exist forother reasons, such as the low commitment of the plant management to quality control.

In other cases, a complying practice might be dependent on no complying practices.This type of relationship indicates that more work than necessary may have been devotedto ensure compliance, since important requirements for compliance are not in place. Thishappens in MC 1 with the practice of one-piece-flow (ONE, full compliance), whichdepends on four other practices that are either in non-compliance or partial compliance.In fact, this means that ONE compliance is fragile and can be difficult to sustain in thelong-term without the support of other practices (e.g., since there is neither TPM nor quicksetups, it is likely that workers will be tempted to produce lot sizes that are bigger thanone). Also, Figure 11 shows that, despite being classified as an end practice,

Figure 9. MC1 with two workers.

LP practices Characteristics1. Teamwork and leadership (not used)

The supervisor coordinated a 40 worker department and he used to not be available to substitute workers in MC1. Performance assessment and reward systems were solely based on individual performance.

5. Standardised work (partially used)

Work standards did not include standard stock and cell layout. There were no audits to check compliance with standards. The remaining attributes were complied with.

15. One-piece-flow (fully used) Every worker carried out all three operations, moving the product from one workstation to another and ensuring one-piece-flow.

Figure 10. Example of how results were compiled and presented to company’s representatives.


Dow

nloa

ded

by [

The

Uni

vers

ity O

f M

elbo

urne

Lib

rari

es]

at 0

0:19

30

Sept

embe

r 20

13

ONE performs an important role supporting two other practices (LSS and VIS). Indeed,one-piece flow prevents layout improvisations that otherwise could be made due toexcessive work-in-process which in turn could be detrimental to the visibility of the cell byworkers.

As to pull production, the company’s representatives recognised in the feedbackmeeting that it was not working as expected and they offered some explanations for this:(a) the PPC department ignores the kanban system and issues production orders which arein conflict with the kanbans; (b) workers feel compelled to follow production orders issuedby PPC rather than kanbans, since the production supervisor (the workers’ immediateboss) is also responsible for the PPC department. In addition to this, Figure 11 highlightsthat the stabilisation required for implementing pull production could be achieved if bothbasic and intermediate practices were in place, such as TPM, TWL, QST and VQC.

Based on Figure 11, it is also possible to identify what the focus should be to enhancethe performance of CI. Even though basic practices that support CI are either partially orfully in place, this seems not to be enough due to the shortcomings of TWL. Indeed,according to the experts, CI very strongly depends on TWL (see Table 1).

Some sequences for implementing improvements might also be derived from Figure 11.For instance, the practices of TWL, CI, QST, PULL, and SPR are not in full complianceand depend on each other. A similar sequence exists among the practices TWL, TPM,PULL and SPR, which stresses the far-reaching impact of TWL on MC1 performance.

Figure 11. Relationship model of LP practices in the context of MC1 performance.


Dow

nloa

ded

by [

The

Uni

vers

ity O

f M

elbo

urne

Lib

rari

es]

at 0

0:19

30

Sept

embe

r 20

13

5. Conclusions

This study presented a framework for assessing the use of LP which has two innovative

characteristics: (a) it focuses on assessing the extent to which LP practices are used at thecell level, unlike previous methods of lean assessment that have focused on lean principles

and/or outputs in the plant as a whole; and (b) it includes a model of the relationships

among lean practices, which helps to put into practice the so-called necessity of

understanding LP from a systemic view.Moreover, the framework includes a set of 18 LP practices and their respective

attributes that might be used in a cell. Guidelines for defining practices, attributes anddevising the relationship model were stated in order to make future adaptations of the

framework to the evolution of LP theory and practice easier. Since these guidelines are

explicit, they add conceptual precision and generalisability to the model, once possible

adaptations will be guided by well-defined principles. Indeed, such guidelines might be

useful for adapting the framework to other sectors rather than the one for which it wasprimarily envisioned (i.e., manufacturing). For instance, lean MCs might take on a

different nature in project-type processes (e.g., construction) and in services, thus

demanding adaptations to the LP practices and attributes.The framework might be used during several stages of the life-cycle of a cell: (a) during

the design of new lean cells, as a check-list and source of insights for designers; (b) as a

supporting tool to assess the systemic impact of design changes after starting celloperations; and (c) as a tool to be applied on a regular basis to several cells within the

plant, providing benchmark measures. Indeed, the application of the framework makes

most sense to companies that are committed to the lean philosophy. It will be difficult, if

not impossible, to support LP practices and attributes in the long-term without the

support of a lean management system permeating the whole organisation.This article also presented a case study of the application of the framework, in order to

illustrate how practical and theoretical insights might be derived from its use as well as to

identify possible shortcomings that may arise when the tool is tested in real-world settings.

In this respect, the case study pointed out that the auditor must be well acquainted with the

concepts of both lean practices and cell attributes, especially because they are defined

qualitatively.With regard to the assessment stages, observing cell functioning proved to be the most

time-consuming. Nevertheless, the observations were essential for a better understanding

of how each practice worked. The interviews were most useful for assessing attributes that

could not be discerned easily and visually, such as that of continuous improvement. The

feedback meeting supported the identification of factors that had an impact on compliance

with the established requirements, since this provided the main opportunity to discuss the

results with company personnel.The assessment of some practices is clearly more demanding than that of others. One of

the reasons for this is the need to understand the relationships between the cell and other

plant departments, since this required data collection is in organisational units other than

the cell being assessed. The assessment of the practice of pull production illustrates this

difficulty, since it was necessary to investigate how the cells exchanged information with

both production planning and control and the purchase department. Equipmentautonomation and layout size and shape also proved to be difficult to assess. In

the former, the difficulty stemmed from the need to have technical knowledge of the

machinery and, for some pieces of equipment, from the need to observe devices inside


Dow

nloa

ded

by [

The

Uni

vers

ity O

f M

elbo

urne

Lib

rari

es]

at 0

0:19

30

Sept

embe

r 20

13

the machine. In the latter, the difficulty was due to the lack of a more precise definition

of the attributes.The case study also indicated that the relationship model of LP practices proved to be

useful for identifying improvement opportunities. However, it was clear that such

identification is not straightforward. Due to the complexity of the model, careful analysis

and an experienced auditor are necessary if practical insights are to be extracted from the

model.Moreover, it is worth reporting that the proposed framework has been successfully

applied in several other real cells, beyond the case study presented in this article. In fact,

the authors of this study have presented the framework as a content item of courses on

lean production offered to practitioners and graduate students. Thus, a number of

evaluators have applied the tool in cells from different industries (e.g., auto parts; shoes;

electronics). Considering the results obtained so far, the proposal has presented good

generalisability.Finally, this study indicated opportunities for future studies, such as: (a) improving the

model of the interfaces among lean practices, based on the opinions of a larger set of

experts; (b) developing a model of the relationships among LP practices and cell attributes;

(c) testing existing methods (e.g., concept maps) or developing new methods for analysing

the interfaces among lean practices and among these and the cell attributes; (d) adapting

the framework to the features of cells and LP in other sectors, such as services;

(e) developing similar frameworks for other process types, such as assembly lines, job-

shops and projects – this will allow assessments of lean practices along complete value

streams; (f) developing a method for assigning weights of importance to the practices,

which could provide additional support for identifying practices that should be prioritised;

and (g) developing and applying criteria for inferring the validity of the framework, such

as its reliability and usability.

References

Black, J.T., 2007. Design rules for implementing the Toyota Production System. International

Journal of Production Research, 45 (16), 3639–3664.Burbidge, J.L., 1996. The first step in planning group technology. International Journal of Production

Economics, 43 (2–3), 261–266.Chandrasekharan, M. and Rajagopalan, R., 1989. GROUPABIL1TY: an analysis of the properties

of binary data matrices for group technology. International Journal of Production Research,

27 (6), 1035–1052.Chiesa, V., Coughlan, P., and Voss, C., 1996. Development of a technical innovation audit. Journal

of Production and Innovation Management, 13 (2), 105–136.Doolen, T. and Hacker, M., 2005. A review of lean assessment in organizations: an exploratory

study of lean practices by electronic manufacturers. Journal of Manufacturing Systems, 24 (1),

55–67.Gurumurthy, A. and Kodali, R., 2009. Application of benchmarking for assessing the lean

manufacturing implementation. Benchmarking: An International Journal, 16 (2), 274–308.

Hendrick, H.W. and Kleine, R.B.M., 2001. Macroergonomics: an introduction to work system design.

Santa Monica, CA: Human Factors and Ergonomics Society.

Hyer, N.L. and Brown, K.A., 1999. The discipline of real cells. Journal of Operations Management,

17 (5), 557–574.


Dow

nloa

ded

by [

The

Uni

vers

ity O

f M

elbo

urne

Lib

rari

es]

at 0

0:19

30

Sept

embe

r 20

13

Hyer, N.L., Brown, K.A., and Zimmerman, S., 1999. A socio-technical systems approach to celldesign: case study and analysis. Journal of Operations Management, 17 (2), 179–203.

Hyer, N. and Wemmerlov, U., 2002. Reorganizing the factory: competing through cellularmanufacturing. New York: Productivity Press.

Karlsson, C. and Ahlstrom, P., 1996. Assessing changes towards lean production. InternationalJournal of Operations and Production Management, 16 (2), 24–41.

Lane, G., 2007. Made-to-order lean: excelling in a high-mix, low-volume environment. New York:

Productivity Press.Lean Enterprise Model, 1998. LAI, lean aerospace initiative [online]. Available from: http://

web.mit.edu/lean/ [Accessed 6 July 2005].

Liker, J., 2004. The Toyota way: 14 management principles from the world’s greatest manufacturer.New York: McGraw-Hill.

Mann, D., 2005. Creating a lean culture: tools to sustain lean conversion. New York: Productivity

Press.Maskell, B. and Baggaley, B., 2004. Practical lean accounting: a proven system for measuring and

managing the lean enterprise. New York: Productivity Press.Miltenburg, J., 2001. U-shaped production lines: a review of theory and practice. International

Journal of Production Economics, 70 (3), 201–214.Miltenburg, J., 2005.Manufacturing strategy: how to formulate and implement a winning plan. 2nd ed.

New York: Productivity Press.

Monden, Y., 1998. Toyota Production System: practical approach to production management. 3rd ed.Norcross, GA: Industrial Engineering and Management Press.

Rother, M. and Harris, R., 2001. Creating continuous flow: an action guide for managers. Cambridge,

MA: Lean Enterprise Institute.SAE International, 1999. [online]. Available from: http://www.sae.org/technical/papers/ [Accessed

25 August 2006].Satoglu, S.I., Durmusoglu, B., and Dogan, I., 2006. Evaluation of the conversion from central

storage to decentralized storages in cellular manufacturing environments using activity-basedcosting. International Journal of Production Economics, 103 (2), 616–632.

Shafer, S.M., Meredith, J.M., and Marsh, R.F., 1995. A taxonomy for alternative equipment

groupings in batch environments. OMEGA: The International Journal of Management Science,23 (4), 361–376.

Shah, R. and Ward, P., 2003. Lean manufacturing: context, practice bundles, and performance.

Journal of Operations Management, 21 (2), 129–149.Sheridan, J., 1990. World-class manufacturing. Industry Week, 239 (13), 36–46.Shingo, S., 1989. A study of the Toyota Production System from an industrial engineering viewpoint.

Portland, OR: Productivity Press.Shingo Prize, 1998. Recognizing business excellence in the United States, Canada and Mexico:

application guidelines [online]. Available from: http://www.shingoprize.org/ [Accessed 22August 2005].

Smalley, A., 2004. Creating level pull. Cambridge, MA: Lean Enterprise Institute.Spear, S. and Bowen, H.K., 1999. Decoding the DNA of the Toyota Production System. Harvard

Business Review, September, 97–106.

Wan, H. and Chen, F., 2008. A leanness measure of manufacturing systems for quantifying impactsof lean initiatives. International Journal of Production Research, 46 (23), 6567–6584.

White, R., Pearson, J., and Wilson, J., 1999. JIT manufacturing: a survey of implementation in small

and large US manufacturers. Management Science, 45 (1), 1–15.Womack, J. and Jones, D., 1998. Lean thinking: banish waste and create wealth in your corporation.

New York: Simon and Schuster.


Dow

nloa

ded

by [

The

Uni

vers

ity O

f M

elbo

urne

Lib

rari

es]

at 0

0:19

30

Sept

embe

r 20

13

Documents

A framework for assessing the use of lean production practices in manufacturing cells