Digital twins are known in production of technical products. Here we explain how it may optimize maritime logistics processes. The presentation was prepared for the LOGMS conference in Singapore 2019

1. Digital Twin
To Improve Efficiency by Forecasting Upcoming Bottle-Necks
Prof. Dr.-Ing. Holger Schuett
akquinet port consulting GmbH
Institute of Shipping Economics and Logistics
University of Applied Sciences Bremerhaven
- Holger.Schuett@akquinet.de

2. Agenda
Digital Twin
to Improve
Efficiency
Simulation - Emulation
Introduction
Digital Twin for Terminals
Applications
Conclusions

3. Agenda
Digital Twin
to Improve
Efficiency
Introduction

4. AKQUINET port consulting
 Prof. Dr. Holger Schütt
 30 years of port, terminals
and automation experience.
 Researcher at Institute of
Shipping Economics and
Logistics
 Professor at University of
Applied Sciences
Bremerhaven
Norbert Klettner
Over 15 years of port,
terminals and TOS experience.
CHESSCON
Some 30 years of simulation,
emulation and optimization
experience.

5. Agenda
Digital Twin
to Improve
Efficiency
Simulation - Emulation

6. Starting in the 90-ies
 Terminal planning and design
strategical planning
9

7. Overall Terminal simulation
 planning of new terminals
 reorganise existing terminals
 layout comparison
 evaluation of handling strategies
 utilisation of equipment
 easy-to-operate user interface
SIMULATION

8. Animation

9. Case study
13
Comparison of operation systems selected
equipment
use
evaluation
production
centres
SC 1 over 3 RTG/TC
No. of STSCs 12 12 12
No. of SCs 45 X X
No. of TCs/AGVs X 53 56
No.of RTGs/RMGs X 25 17
STSC operation hours 1130 1074 1057
SC operation hours 5016 X X
TC/AGV operation hours X 5683 5300
RTG/RMG operation hours X 3141 2737
aver. service time external trucks 20 min 8 min 4 min
average service time 13.6 12 11.7
DS1000 aver. moves/hr (total) 147.0 167.0 171.0
aver. moves/hr per STSC 29.5 32.3 33.4
average service time 12.5 10.5 10.1
DS800 aver. moves/hr (total) 128.0 152.0 158.0
aver. moves/hr per STSC 29.3 31.5 32.9
average service time 4.5 4.3 4.1
F120 aver. moves/hr (total) 53.0 56.0 59.0
aver. moves/hr per STSC 21.3 21.6 22.83
average service time 8.8 8.0 7.8
F250 aver. moves/hr (total) 57.0 62.33 64.0
aver. moves/hr per STSC 20.4 21.5 22.6
total berth operation time 218.0 195.0 189.0
costs per move [€] 143.36 157.44 132.76
RMG/AGV
auto
costs
The decision from an economical view is supported
based on operational costs and investment
... But what are the
ecological impacts
of the terminal?

10. CAPACITY
Input
 vessel plan with differentiation by vesseltypes
 quay layout, slots per ct type
Output
 determination of quay capacity
 analysing stacking area‘s capacity
 number of STS cranes required
 Workload on stacking areas, waterside, gate and
rail

11. Post - Millenial
 Terminal Start-Up and Operation
tactical planning
16

12. Coupling simulation to the TOS
 optimise utilisation of equipment
 improve your terminal productivity
 optimise handling strategies
(fine-tuning TOS)
 reduce operational costs
 increase the skill of your control staff
(TOS training for control staff)
VIRTUAL TERMINAL

13. 18
Virtual Terminal

14. Nowadays
 Forecasting the coming operation
operational planning
20

15. Fast simulation for forecasting
 verify your current shift planning
 based on actual TOS data
 high speed simulation – about 3000
moves/minute
 improve yard allocations
 detect yard clashes before the shift
begins
SHIFT PREVIEW
Development funded by

16. Shift Preview
0 step:
day to day work
use the TOS
to plan the next shift
1 step:
shift planning finished

17. Shift Preview 2nd step:
Import planning state
automatically
3rd step:
Running fast simulation
of the operation

18. Shift Preview
4th step: intuitive evaluation of
the efficiency

19. Comparison
Scenario Decisions Level of
detail
Speed Application
Simulation to be defined internal light
TOS
low - medium very high strategic
Emulation defined in
TOS
real TOS very high low (due to
TOS-coupling)
tactical
Preview downloaded
from TOS
internal light
TOS using
real TOS
parameters
high high (operational)

20. Agenda
Digital Twin
to Improve
Efficiency
Digital Twin for Terminals

21. A digital twin
 … is a digital representation of a real-world entity or system. (Gartner IT)1
 … is a virtual representation of a physical product or process, used to understand
and predict the physical counterpart’s performance characteristics. 2
(Performance Digital Twins: Using digital twins capture, analyze, and act on
operational data)
 The concept and model of the Digital Twin was publicly introduced in 2002 by Dr.
Michael Grieves, then of the University of Michigan, at a Society of Manufacturing
Engineers conference in Troy, Michigan[20]. The concept which had a few different
names was subsequently called the Digital Twin by John Vickers of NASA in a 2010
Roadmap Report[21]. The Digital Twin consists of three parts: The physical
product, the virtual product, and the connection between the two products.
Definition
1: https://www.gartner.com/it-glossary/digital-twin/
2: https://www.plm.automation.siemens.com/global/en/our-story/glossary/digital-twin/24465
3: https://en.wikipedia.org/wiki/Digital_twin

22. TOS
Real
(physical product)
Digital Twin
To build the digital Twin
 use the emulation model
 connect it to the TOS (listen only) ( Live View)
 use Shift Preview module to forecast operation
Virtual
(listen only
drop off
pick up
position
…)

26. Comparison
Scenario Decisions Level of
detail
Speed Application
Simulation to be defined internal light
TOS
low - medium very high strategic
Emulation defined in
TOS
real TOS very high low (due to
TOS-coupling)
tactical
Preview downloaded
from TOS
internal light
TOS using
real TOS
parameters
high high (operational)
Digital Twin permanently
updated by
TOS
internal light
TOS using
real TOS
parameters
high high operational

27. Agenda
Digital Twin
to Improve
Efficiency
Applications

28. Calibrating the simulation model by analysing the behaviour of the real world
- listen to the real world
- analyse the behaviour and find changes
- evaluate parameters of the real objects (e.g. productivity of manned devices 
learning curve)
- adapt these changes to the emulators of the objects to be more precise for
future simulation analysis
Applications

29. Re-act on exceptions happening in the real world
- exception occurs (e.g. crane breakdown)
- planner has to react very fast (e.g. define new crane split)
- test new strategy by fast simulation (Preview)
- potential bottle-necks will be shown
- planner will pro-actively (before they occur) optimise his strategy
Applications

30. Automated check of the current plan
- preview module is automatically started in predefined intervals
- predefined thresholds are compared to the simulation results
- productivity/efficiency lags may be identified in advance
- check of productivity figures (e.g. planned end of operation at the vessel)
- differences may be automatically sent to the planner/shown on the screen
Applications

31. Agenda
Digital Twin
to Improve
Efficiency
Conclusions

32. Digital Twins at container terminals may be
- built by using the emulation model of a terminal
- connected to the real terminal by listening to the TOS messages
- use the preview module (fast simulation) to predict the future behaviour
They combine
- the detailled behaviour of the emulators
- the current state of the terminal (by listening to the messages)
- the fast simulation of the preview module
and thus may be used for improving the operational planning.
Conclusions

33. Already existing
- emulation models
- preview module based on the emulators
- interface to the TOS messages and synchronisation at program start
- OEE evaluation for fast analysis if simulation results
To Dos
- handle different actions of real and virtual objects (e.g. arrival of transport
equipment at destination)
- analysing changes in behaviour (learning curves) by means of AI
Current state and still To-Dos