Applications of Simulation-based Analysis by Digital Twins for Next Generation Ports

1

Dr. LI HaobinNational University of SingaporeDepartment of Industrial Systems Engineering and ManagementCentre of Excellence in Modelling and Simulation for Next Generation Ports

Outline

Framework for Simulation-based AnalysisSimulation Evaluation

Simulation based Optimization

Hybrid Analysis

Case StudiesHierarchical Modelling for Mega Container Port

Capacity Planning for Container Terminals

Capacity Assessment for General Cargo Port

Yard Planning for Transshipment Terminals

Configuration Optimization for AGV Markers

Yard Block Reshuffling Analysis

2

3

Framework for Simulation-based Analysis

Big Data Analytics / Deep Learning

Optimal Configuration

SIMULATION ANALYTICS

Random Seed

SIMULATION IN OPTIMIZATION

Sim.-based Optimization (Search Algorithm / Ranking & Selection)

Event 1

Event 2

Event 3

Event 1

Executor

Entity 1

Entity 3

Entity 2

Load 1

Load 2

STATE

EVENTS

ConfigurationEn

viro

nmen

t /

Initi

al S

tate

Obj

ectiv

es

Event 1

Event 2

Event 4

Event 5

Event 3

Event 3(OPT.)

Event 1

Event 2

Event 4

Event 5

OPTIMIZATION

OPTIMIZATION IN SIMULATION

Conditions

ASSETS（ Static（

SANDBOX（ Dynamic（

Seamless Integration betweenSimulation and Optimization

4

Framework for Simulation-based Analysis

Simulation Evaluation

Event 1

Event 2

Event 3

Event 1

Executor

Entity 1

Entity 3

Entity 2

Load 1

Load 2

STATE

EVENTS

ConfigurationEn

viro

nmen

t /

Initi

al S

tate

Obj

ectiv

es

Event 1

Event 2

Event 4

Event 5

Event 3

Event 3(OPT.)

Event 1

Event 2

Event 4

Event 5

OPTIMIZATION

OPTIMIZATION IN SIMULATION

Conditions

ASSETS（ Static（

SANDBOX（ Dynamic（

Evaluation of a given system configuration

Evaluation of operational rules / optimization logicsin dynamic and stochastic circumstances

Berth Allocation / Scheduling

Yard Allocation

Vehicle Dispatching / Dynamic Routing

1

2

5

Framework for Simulation-based Analysis

Simulation based Optimization

Random Seed

SIMULATION IN OPTIMIZATION

Sim.-based Optimization (Search Algorithm / Ranking & Selection)

Event 1

Event 2

Event 3

Event 1

Executor

Entity 1

Entity 3

Entity 2

Load 1

Load 2

STATE

EVENTS

ConfigurationEn

viro

nmen

t /

Initi

al S

tate

Obj

ectiv

es

Event 1

Event 2

Event 4

Event 5

Event 3

Event 3(OPT.)

Conditions

ASSETS（ Static（

SANDBOX（ Dynamic（

Optimization for maximum / minimum objective valuesby adjusting system configurations

Port Design

Equipment Selection

Capacity Planning

Ranking and SelectionFrom a given set of alternatives

Large Scale SearchFrom an infinite / large scale set

3

4

6

Framework for Simulation-based Analysis

Hybrid Analysis

Hybrid Optimization for objective valuesby Evaluating and adjusting operational rules

Port Design

Equipment Selection

Capacity Planning

Random Seed

SIMULATION IN OPTIMIZATION

Sim.-based Optimization (Search Algorithm / Ranking & Selection)

Event 1

Event 2

Event 3

Event 1

Executor

Entity 1

Entity 3

Entity 2

Load 1

Load 2

STATE

EVENTS

ConfigurationEn

viro

nmen

t /

Initi

al S

tate

Obj

ectiv

es

Event 1

Event 2

Event 4

Event 5

Event 3

Event 3(OPT.)

Event 1

Event 2

Event 4

Event 5

OPTIMIZATION

OPTIMIZATION IN SIMULATION

Conditions

ASSETS（ Static（

SANDBOX（ Dynamic（

5

7

Case Studies

Hierarchical Modelling for Mega Container Port

MEGA CONTAINER PORT

O2DES.NetA Discrete-Event Digital Twin

for

9

Case Studies

Capacity Planning for Container Terminals

Definition of Capacity Assessment for Port Terminals

Given a configuration of the port terminals, the capacity assessment is to find the maximum throughput it is able

to handle, respect to the constraints that,

1) the BoA rate shall be above a threshold, and/or

2) the utilization of port facility shall be below a threshold,

in which, the BoA rate and utilization of port facilities shall be evaluated by a Discrete-Event Simulation.

Example:capacity assessment for cement

terminals by considering BoA rate and berth utilization (Li et al. 2017)

10

Case Studies

Capacity Assessment for General Cargo Port

• General cargo ports are very unique in comparison to the container ports, in terms of

• variety of cargo types, and the respective equipment that handles the cargo;

• mix of different cargo types sharing the same pool of port facilities;

• the pattern of vessel arrivals (bus mode vs. taxi mode), which plays an important role in BoA rate;

• The studies of capacity assessment on container ports cannot be directly applied to the general cargoports.

• The existing study on the general cargo ports has focus only on an isolated single cargo type, which isinsufficient.

11

Case Studies

Capacity Assessment for General Cargo Port

Entity-Relationship Diagram for Cargo Mix and Handling Resources in General Cargo Port

12

Case Studies

Yard Planning for Transshipment Terminals

Generation of Transshipment

MatrixDecide Minimum

Cluster Size

For each transshipment pair (O/D vessels), which are not

fully allocated

Filter out infeasible clusters (e.g., workload threshold)

Identify the cluster with best criteria value (e.g., min.

travelling distance)

Itera

te in

rou

nd-r

obin

/ ra

ndom

ized

seq

uenc

e

Evaluate KPIs

Algorithm Parameters Cluster size Allocation sequence Filtering and sorting conditions

Algorithm KPIs Spread if yard-crane work loads Travelling distances Computational efficiency

Rev

ise

algo

rith

m p

aram

eter

13

Case Studies

Optimization for AGV Markers Configuration

Generation of AGV Trajectories(Scenarios)

Select Markers Pattern and Density(Decisions)

Simulate for each scenario

Observe performance measure (i.e., max. inter-reading distance,

and ratio of duplicates)

Measurement for robustness Adj

ust

desi

gn p

atte

rn /

dens

ity

Itera

te

Scen

ario

s

Feasible / Optimal Design

0

10

20

30

40

50

60

70

0 500 1000 1500 2000 2500 3000

Perc

enta

ge o

f Res

huffl

ings

# Containers

6-Tiers Low 6-Tiers Medium6-Tiers High 9-Tiers Low9-Tiers Medium 9-Tiers High

14

Case Studies

Yard Block Reshuffling Analysis

010203040506070

0 1 2 3 4 5 6

Perc

enta

ge o

f Res

huffl

ings

Operational Stacking Height (Avg. before Loading)

010203040506070

0 1 2 3 4 5 6

Perc

enta

ge o

f Res

huffl

ings

Operational Stacking Height (Avg. before Loading)

010203040506070

0 1 2 3 4 5 6

Perc

enta

ge o

f Res

huffl

ings

Operational Stacking Height (Avg. before Loading)

500-Slots x 6-Tiers

0

10

20

30

40

50

60

70

0 1 2 3 4 5 6 7 8 9

Perc

enta

ge o

f Res

huffl

ings

Operational Stacking Height (Avg. before Loading)

0

10

20

30

40

50

60

70

0 1 2 3 4 5 6 7 8 9

Perc

enta

ge o

f Res

huffl

ings

Operational Stacking Height (Avg. before Loading)

010203040506070

0 1 2 3 4 5 6 7 8 9

Perc

enta

ge o

f Res

huffl

ings

Operational Stacking Height (Avg. before Loading)

High-Variety(500 Ccategories)

Low-Variety(300 Ccategories)

Medium-Variety(400 Ccategories)

334-Slots x 9-Tiers

Space for 3,000 TEUs

15

A Simulation-based Capacity Assessment Toolbox For Jurong Port

Thank you!

Q&A