Using dynamic modelling tools - aapa-ports.org€¦ · Using dynamic modelling tools Dr. ir. Yvo Saanen (c) TBA 2007 / Semi-automated terminals / AAPA 2007 2 1. Automated terminal

Designing semiDesigning semi--automated terminalsautomated terminalsUsing dynamic modelling tools Using dynamic modelling tools

Dr. Dr. irir. Yvo Saanen. Yvo Saanen

(c) TBA 2007 / Semi(c) TBA 2007 / Semi--automated terminals / AAPA 2007automated terminals / AAPA 2007

2

1.1. Automated terminal amid their main competitorsAutomated terminal amid their main competitors

2.2. Design approach (semiDesign approach (semi--) automated terminal) automated terminal

3.3. Emulation as tool to ensure software qualityEmulation as tool to ensure software quality

4.4. Concluding remarksConcluding remarks

Contents of presentationContents of presentation

Automated terminals amid their main competitorsAutomated terminals amid their main competitors

11


7

The most efficient (semiThe most efficient (semi--) automated concept nowadays) automated concept nowadays4. RMG + AGV4. RMG + AGV

§ Density: 1500 TEU / ha§ Terminal capacity: 2000 TEU / m quay§ Operational cost reference number: 1§ Investment cost reference number: 5


8

Why automate?Why automate?

Designing a semiDesigning a semi--automated container terminalautomated container terminal

22


10

Simulation approach for robust design of an automated terminalSimulation approach for robust design of an automated terminal

§§ S.M.A.R.T. definition of goals and KPIS.M.A.R.T. definition of goals and KPI‘‘ss• Realistic operational scenarios• Performance targets• Operational costs / investment

§§ Static terminal design:Static terminal design:• Berth capacity• Storage capacity

§§ SimulationSimulation of of quayquay operationsoperations

„„setting the scenariossetting the scenarios““

§§ Initial drawings of alternativesInitial drawings of alternatives

§§ SimulationSimulation (alternative) terminal operations(alternative) terminal operations

„„productivity assessmentproductivity assessment““• Equipment selection• Dimensioning of „handling system“• Defining the logistical concept• Defining business logic

§§ Simulation of terminal during 6 weeksSimulation of terminal during 6 weeks

§§ TerminalTerminal costcost calculationcalculation ((investmentinvestment and and

operational)operational)

§§ Implementation planImplementation plan

Inputdata

Area constraintsGoals & requirements

Static terminaldimensioning

Preliminaryterminal design

Quay wallsimulation

Final principalterminal design

Handling systemdesign

Cost calculation

Simulationtools

Cost model

Productivity calc.and assessmentRisk assessment

Satisfied?

No

Yes

Project continuation

Handling systemsimulation

Inputdata

Area constraintsGoals & requirements

Static terminaldimensioning

Preliminaryterminal design

Quay wallsimulation

Final principalterminal design

Handling systemdesign

Cost calculation

Simulationtools

Cost model

Productivity calc.and assessmentRisk assessment

Satisfied?

No

Yes

Project continuation

Handling systemsimulation


11

Simulation cycle (in each phase)Simulation cycle (in each phase)

model“as is”

models“to be”

currentsituation

newsituation

search forsolutions

pre-evaluation

diagnoseproblemvalidation

problem identificationand specification

choice andimplementation

post-evaluation

evaluation

model“as is”

models“to be”

currentsituation

newsituation

search forsolutions

pre-evaluation

diagnoseproblemvalidation

problem identificationand specification

choice andimplementation

post-evaluation

evaluation


12

The key: balanced design of the yardThe key: balanced design of the yard


13

Right balance / right designRight balance / right design

§§ Target productivity waterside Target productivity waterside –– capable of handling future demands?capable of handling future demands?• Peak loads (twin lift, tandem 40, dual cycling, reefer handling, MTs)• How many QCs on one vessel?

§§ Peak at the landside, gate + rail?Peak at the landside, gate + rail?

§§ Typical peak scenario: how many out of all QCs are actually in oTypical peak scenario: how many out of all QCs are actually in operation?peration?

§§ How many stacking cranes to meet this demand, in relation to theHow many stacking cranes to meet this demand, in relation to their specifications:ir specifications:• Length of stack module• Gantry speed & acceleration• Hoist speed and acceleration• Design of the waterside interchange zone (“buffering”)• Rolling percentage (effective rehandling)• Landside operation (remote operation)• Safety rules (e.g. access to interchange zones)

§§ Which automated transportation system is the most cost efficientWhich automated transportation system is the most cost efficient??

§§ How much waterside equipment to meet demands?How much waterside equipment to meet demands?• Interaction between yard equipment and transportation equipment• Sequence during vessel loading• Etc.


14

Right design / which stack handling system?Right design / which stack handling system?

§§ The Twin RMG:The Twin RMG:

• 2 similar RMGs on one track

• Ability to mutually support

§§ The crossThe cross--over twin RMG:over twin RMG:

• One large, one smaller RMG on 2 tracks

• Ability to pass, and work on either side

§§ The crossThe cross--over tri RMG:over tri RMG:

• One large, two smaller RMG on 2 tracks

• One small RMG for either side

• Ability to pass, and work on either side


15

Right design / which transportation system?Right design / which transportation system?


16

Various combinations of yard and transportation equipmentVarious combinations of yard and transportation equipment(example)(example)

Iso performance / cost graphsquay length = 400m, 4 quay cranes, pooled vehicles

6

7

8

9

10

11

12

13

14

4 5 6 7 8 9 10 11 12 13 14 15 16# Transport vehicles (ShCs)

# St

acki

ng c

rane

s

Berth productivity 100 moves/hour Berth productivity 110 moves/hourBerth productivity 120 moves/hour Berth productivity 130 moves/hour

32

34

37

39

42

47


17

Various terminal designs Various terminal designs depending on local circumstancedepending on local circumstance

§§ Terminal 1:Terminal 1:

§§ 2.0 M TEU2.0 M TEU

§§ 5% transhipment5% transhipment

§§ Dwell time: 9 daysDwell time: 9 days

§§ Peak productivity (WS/LS): 550/400Peak productivity (WS/LS): 550/400

§§ Configuration: 30 modules, 60 TEU long, 8 wideConfiguration: 30 modules, 60 TEU long, 8 wide


§§ 3.5 M TEU3.5 M TEU






§§ 4.5 M TEU4.5 M TEU


§§ Dwell time: 4.5 daysDwell time: 4.5 days




§§ 1.8 M TEU1.8 M TEU




§§ Configuration: 24 modules, 26/52 TEU long, Configuration: 24 modules, 26/52 TEU long,

11/8 wide11/8 wide


18

ASC specificationASC specification

Effect of the stacklength2-8 Corridor - Vgantry = 4,0 m/s - Agantry = 0,3 m/s2 - Vtrolley = 1,0 m/s - Atrolley = 0,3 m/s2 - Vhoist

= 1,0 m/s - Ahoist = 0,33 m/s2 - Scenario 3

19.8 19.418.6 18.5 17.9

17.3

15.2 14.6 14.1 13.9 13.4 13.2

0

5

10

15

20

25

30 35 40 45 50 55 60Stack length [TEU]

AC

S pr

oduc

tivity

[mph

]

Waterside crane

Landside crane

Effect of gantry speedAgantry = 0.3 m/s2 - Vtrolley = 1,0 m/s - Atrolley = 0,3 m/s2 - Vhoist = 1,0 m/s - Ahoist = 0,33 m/s2

21.4 22.0 22.0 21.9

14.8 15.5 15.8 15.8 15.9

21.7

0

5

10

15

20

25

30

3.0 3.5 4.0 4.5 5.0

Gantry speed (m/s)

AC

S pr

oduc

tivity

[mph

]

Waterside craneLandside crane


19

Design of the reefer solutionDesign of the reefer solution

§§ Reefer solution:Reefer solution:

• Dependent on reefer share

• Dependent on # reefer mechanics

• Dependent on reefer solution (location in

the stack module, access, ASC restrictions)

Antwerp Gateway Project - Net quay crane productivitySmall terminal - 8 stack modules - 90 LS moves - 4 QCs - 3 SC per QC

35.433.2

27.4

35.3

31.8

24.7

0

5

10

15

20

25

30

35

40

45

50

0% 10% 20% 0% 10% 20%

1:4 rf mod - 20 cntrs/bay - 2 rf mechs 1:4 rf mod - 20 cntrs/bay - 1 rf mechsPercentage reefers of total containers

Net

QC

pro

duct

ivity

(bx/

h)Antwerp Gateway Project - Net quay crane productivity

Small terminal - 8 stack modules - 90 LS moves - 4 QCs - 3 SC per QC

35.434.2

29.7

35.8

32.6

26.3

0

5

10

15

20

25

30

35

40

45

50

0% 10% 20% 0% 10% 20%


Net

QC

pro

duct

ivity

(bx/

h)

Antwerp Gateway Project - Net quay crane productivitySmall terminal - 8 stack modules - 90 LS moves - 4 QCs - 3 SC per QC

35.5 34.833.5

35.6

33.4

27.8

0

5

10

15

20

25

30

35

40

45

50

0% 10% 20% 0% 10% 20%


Net

QC

pro

duct

ivity

(bx/

h)

1 out of 4 blocks with reefers 1 out of 3 blocks with reefers

1 out of 2 blocks with reefers


20

Traffic analysisTraffic analysis

X 650 850 1050 1250 145040

56

72

88

104

120

x coordinates

DP World - London Gateway - Traffic density22 QCs - 88 lift AGVs - correct berthing

Right side of terminal - average number of vehicles per hour

0-15 15-30 30-45 45-60 60-75 75-90 90-105 105-120 120-135 135-150 150-165

165-180 180-195 195-210 210-225 225-240 240-255 255-270 270-285 285-300 300-315 315-330

4 QC vessel6 QC vessel

stack 1stack 29

Traffic density


21

Performance under breakPerformance under break--down circumstancesdown circumstances(example)(example)

Truck service time module A257

-3600

0

3600

7200

10800

0 3600 7200 10800 14400 18000 21600 25200

Time of arrival at gate (s)

Tim

e (s

)

Waiting time for transfer point Terminal time Breakdown Breakdown resolved


22

Typical deliverables of a design studyTypical deliverables of a design study

§§ Berth simulation:Berth simulation:• Berth occupancy• Quay crane utilization• Vessel service times• Yard occupation throughout the year• Handling peaks waterside & landside

§§ Handling system simulation:Handling system simulation:• Required numbers of yard cranes & shuttle

carriers• Optimized yard design• Reefer solution• Design of interchange zones• Effect of breakdown and mitigation

strategies• Algorithms for dispatching, scheduling and

grounding

§§ Other:Other:• Solution (conceptually) for scheduling /

dispatching ASCs• Solution (conceptually) for scheduling /

dispatching ShCs

QC productivity - Vehicle Comparison [6 QCs @ 45 ccph, 25 TwinRMG modules, 350 landside bx/h]

0

5

10

15

20

25

30

35

40

45

50

9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54

Total number of vehicles available

Prod

uctiv

ity (b

x/hr

)AGVShCAGV_LiftALVC-AGV

RMG: Productive moves all cranes

22.6

25.4

28.0

23.9

26.2

23.9

27.1

29.9

32.8

34.8

36.1 36.2

30.7

35.034.6

33.3

33.533.1

31.8

30.728.2

20

25

30

35

40

0 3 6 9 12 15 18Number of container moves landside

Prod

uctiv

e m

oves

per

hou

r all

cran

es

Twin

Dual (Large LS+WS)

Tri (Large LS+WS)

100% 85% 70% 55% 40% 25% Transhipment ratio


23

Typical deliverables of a design studyTypical deliverables of a design study

Emulation as a way to ensure software quality and performanceEmulation as a way to ensure software quality and performance

or

33


25

Link software to a virtual terminal (CONTROLS)Link software to a virtual terminal (CONTROLS)

TOS (control system)TOS (control system)TOS (control system) TOS (control system)TOS (control system)TOS (control system)

Real terminalReal terminal Virtual terminalVirtual terminal


26

Key properties of Key properties of dynamic modelsdynamic models

1.1. DynamicsDynamics

2.2. Safe and inexpensive trial and error environmentSafe and inexpensive trial and error environment

3.3. Analysis of nonAnalysis of non--repetitive eventsrepetitive events

4.4. See processSee process

5.5. Quantify and prioritizeQuantify and prioritize


27

ÉÉ TrainTrain ““control roomcontrol room”” operators:operators:

• Operate a virtual terminal

• No “learning” impact to operation

• Get immediate feedback

• Practice on irregular operations

3T3T--Concept: Concept: TTesting, esting, TTuning and uning and TTrainingraining

ÅÅ TestTest new software:new software:

• Reducing risk

• More focus on performance

• Knowing problems earlier in process

• Insight for non software experts

ÇÇ TuneTune your operation:your operation:

• Anticipation on problems

• Validation of planning

• Replay of operation

• Running future growth scenarios

Key references:Key references:

§§ Pusan Newport (Zodiac)Pusan Newport (Zodiac)

§§ MSC Home (SPACE/TRAFIC)MSC Home (SPACE/TRAFIC)

§§ APMT Virginia (SPARCS)APMT Virginia (SPARCS)

§§ EurokaiEurokai Hamburg (TOPHamburg (TOP--X)X)

§§ APMT Rotterdam (SPACE/TRAFIC)APMT Rotterdam (SPACE/TRAFIC)

§§ Euromax (SPARCS / TEAMS)Euromax (SPARCS / TEAMS)

§§ APMT Aarhus (SPARCS)APMT Aarhus (SPARCS)


28

Example: testing the TOSExample: testing the TOS

Concluding remarksConcluding remarks

or

44


30

ConclusionsConclusions

§§ With the right approach, and the proper tools, automated terminaWith the right approach, and the proper tools, automated terminals can be ls can be

designed and implemented:designed and implemented:

• Without major risk

• Within time

• Within budget

• Delivering the targeted productivity

§§ The design will very from site to site, as many conditions deterThe design will very from site to site, as many conditions determine the mine the ““optimaloptimal””

designdesign

§§ Automated terminals Automated terminals –– more than any other type of terminal more than any other type of terminal –– require proper require proper

planning, with a long term vision on service levels and handlingplanning, with a long term vision on service levels and handling capabilitiescapabilities


32

Contact detailsContact details

TBA B.V.TBA B.V.

Dr. Yvo Saanen Dr. Yvo Saanen –– Managing DirectorManaging Director

VulcanuswegVulcanusweg 259259

2624 AV DELFT2624 AV DELFT

The NetherlandsThe Netherlands

Tel.Tel. +31 15 3805775+31 15 3805775

Fax.Fax. +31 15 3805763+31 15 3805763

WebWeb www.tba.nlwww.tba.nl

EE--mailmail [email protected]@tba.nl

Documents

Using dynamic modelling tools - aapa-ports.org€¦ · Using dynamic modelling tools Dr. ir. Yvo Saanen (c) TBA 2007 / Semi-automated terminals / AAPA 2007 2 1. Automated terminal