port multimodal inland mode of transportation predictor ... - …€¦ · the proposed tool, aims...

Minisymposium: Mathematics of Logistics: emerging trends in Optimization and Simulation modelling

Port Multimodal Inland mode of transportation predictor & prescriptor

Stefano Persi1*, Carlos Morillo2 1. Mosaic Factor, Spain, [email protected] 2. Mosaic Factor, Spain, [email protected]

Abstract The steady increase referred to the goods moved by containers in the different ports of Europe, as well as all around the world, the volume of TEU’s has augmented 20% in the last 10 years, has implied an increase of the inland transportation that has posed economic, social and environmental challenges to ports, and their hinterland. To tackle this issue, one of the innovative tools that the European Commission’s project, COREALIS, is developing, is the Cargo Flow Optimizer (CFO) tool, that uses the data available from the key port stakeholders (Port Authority, Terminal Operators, inland transportation operators, etc.) to create a predictive and prescriptive model in order to improve multimodality. The proposed tool, aims to improve the port operations while at the same time it reduces the current negative impacts in the surrounding area. Keywords: MULTIMODALITY, LOGISTICS, PREDICTION

1. Introduction Ports are essential for the European economy; 74% of goods exported or imported to the EU are transported via its seaports. At the same time, the challenges they face are only getting greater: Volumes of cargo increase while they also arrive in a shrinking number of vessels: Post-Panamax vessels have a capacity of more than 18.000 containers. Port operators need to comply with increasingly stricter environmental regulations and societal views for sustainability. A sustainable land-use strategy in and around the port and a strategic transition to new, service-based, management models that improve capacity and efficiency are paramount. They are key enablers for ports that want to keep pace with the ocean carriers needs and establish themselves as trans-shipment hubs with a ‘societal license to operate’; for ports whose land strategy, hinterland accessibility and operations are underpinned by circular economy principles. In this sense, the European Union (EU) is putting significant effort in order to improve and optimize the efficiency of the European ports. To achieve this goal, a number of European projects have been funded within the Ports of the Future program, with COREALIS being one of them. COREALIS proposes a strategic, innovative framework, focused on disruptive technologies, including IoT, data analytics, next generation traffic management and 5G, for modern ports to handle future


capacity, traffic, efficiency and environmental challenges. Through COREALIS, the port will minimize its environmental footprint to the city as well as decrease disturbance to local population through a reduction in the congestion around the port. To tackle these challenges, one of the innovative tools that the COREALIS project will develop is the cargo flow optimizer (CFO) tool that develop a model based on data available from the key Port stakeholders (Port Authority, Terminal Operators, Inland Transportation Operators, etc.) that allows to predict and prescript different solutions related to the port operations as well as to the multimodal transport.

2. Main issues related to port multimodality operations It is already known that one of the basic problems related to the ports and their inland connectivity is referred to the trucks, as the most common inland mode of transport, and the different issues that arise with that mode all along the port and its surroundings. Nowadays, the most commonly used mode of transportation is road transportation. Several reasons make this mode the best fitted for all the carriers:

• Road transportation allows a much wider schedule flexibility due to its main dependence to the driver and its almost full independence to any other issue (shared transportation, rigid timetables, etc.)

• Road transportation allows to move the cargo independently of the number of containers required by the shipper

• Finally, road transportation does not depend on any prioritizing rule that any other transportation mode can apply – the most common case is the one related to rail transportation where passengers are normally prioritized in front of goods

Although for carriers road transportation is often the best solution, this option has different problems that make this solution one to be, at least, mitigated and adjusted to the real needs of all the stakeholders:

• The port detects that a high number of heavy vehicles in the roads of the installation, implies a reduction of the productivity and an increase of congestion and the appearance of a bottleneck

• The terminal and specifically its internal operations, due to the bottleneck that arises with the congestion, can be affected implying a relocation in-real-time of the containers and a reduction of the productivity

• Finally, in the surroundings of the ports different issues related to road inland transportation appear: traffic congestion, atmospheric and acoustic contamination, increase of accident rate, visual impact, etc. These side effects are increased due to, normally, next to the port is located an urban area and, implying that the negative effect of the different impacts affect economically, socially and environmentally to the mentioned urban area.

In fact, one of the key reasons not to use other inland transportation modes is related to the lack of information of these alternatives as well as the option to access, as customers, to these options independently of the total volume to be transported.


3. Scope of study To analyze the issue, we decided to work with the Port of Antwerp. This port is located next to the most populated city of Belgium (and the 2nd metropolitan area behind Brussels) and was in 2017 the 2nd container port of Europe behind Rotterdam (and the number 17 in all the world in 2014) with almost 9 Million TEU per year.

Figure 1: Port of Antwerp

The selection of the port of Antwerp (PoA) was motivated for many reasons:

• PoA is one the biggest ports in Europe and it is strategically located about 88 kilometers from the North Sea at the upper end of the tidal estuary of the river Scheldt

• The city of Antwerp, with more than half a million inhabitants is next to the port

• The roads that are going from and to the port are saturated with trucks involving Antwerp as the most traffic saturated urban area in Belgium and one of the most congested in all Europe

• The location of the Port allows not only to use rail and road mode but also barges to transport the goods inland

'Á

r

N49

Zel

zate

/ K

nokk

e

A12 Brussel(via Boom)

E17 Gent(F) Lille, Paris

A4 (NL)(via Bergen op Zoom)

E19 Brussel / Mons (F) Paris

E314 (D

) Aach

en, K

ölnE

313 Hasselt / L

iègeE

34 (NL

) Ein

dh

oven /(D

) Ru

hrg

ebiet

E19 (N

L) R

otterd

am (via B

reda)E

313 Hasselt / L

iègeE

34 (NL

) Ein

dh

oven /(D

) Ru

hrg

ebiet

R 2

R 2

R 1

E 19

E 19

E 34

E 17

E 34

A 12

KIEL

LILLO

EKEREN

DEURNE

MERKSEM

BERCHEM

WILRIJK

MELSELE

STABROEK

HET ZUIDKRUIBEKE

ZANDVLIET

SINT-ANNA

KLEINE BAREEL

BORGERHOUT

BERENDRECHT

ZWIJNDRECHT

LINKEROEVER

KANAALDOK B1/B2

WAASLANDHAVENOOST

WAASLANDHAVENZUID

WAASLANDHAVENNOORD

5

3

6

7

8

9

1

2

5 4

11

12

13

15

16

5a

17

16

13

12

11

10

LEUGENBERG

Marshall-dok

Kat

tend

ijkdo

k

Noordelijkinsteekdok

Zuidelijkinsteekdok

Insteekdok 1

Insteekdok 4

Insteekdok 3

Insteekdok 2

Lobroekdok

Asi

adok

Houtdok

Straats-burgdok

Industriedok

Duw

vaart

Schuildok

Noordkasteel-dok

Kempischdok

DD 1

DD 8DD 9

DD 5

Schelde - Rijnkanaal

Albertkanaal

Galgenweel

Noordkasteelroeivijver

Put van Fien

Hansadok

Kanaaldok B2

Waaslandkanaal

Kanaaldok B1

Doeldo

k Deur

ganc

kdok

Vras

ened

ok

Churchilldok

Delwaidedok

Kanaaldok B

3

5e Havendok

6e Havendok

Leopolddok

Verre

broe

kdok

Albertdok

4e Havendok

Amerikadok

3e Havendok

2e Havendok

ZANDVLIETSLUIS

KALLOSLUIS

BERENDRECHTSLUIS

BOUDEWIJNSLUIS

VAN CAUWELAERTSLUIS

KATTENDIJKDOKSLUIS

ROYERSSLUIS

Schelde

KIELDRECHTSLUIS

Kennedytunnel

Beve

rent

unne

l

Waaslandtunnel

Tijsmanstunnel

Voetgangerstunnel

Lief

kens

hoek

tunn

el

Lief

kens

hoek

spoo

rver

bind

ing

Ketenisbrug

Prosperbrug

ADPOLLH

GLOBAL CONTAINERSERVICES

8181

81

MSC PSAEUROPEAN TERMINAL

(MPET), K1742

Hobokense polder

Prosperpolder Noord

Haasop

Het Rot

Grote Geule

Blokkersdijk

Groot Buitenschoor

Bospolder

Sint-Annabos

Verrebroekse Plassen

Groot Rietveld

Ekerse Putten

Vlietbos

Doelpolder Noord

Kuifeend

Drijdijck

Burchtse Weel

Melkader

MIDA-zone

Rietveld Kallo

Opstalvalleigebied

Putten Weide

Grote Kreek

Ketenisseschor

Putten West

Galgeschoor

gedempt Doeldok

Fort Sint-Filips

Brakke Kreek

De Zouten

Fort Sint-Marie

Spaans fort

Zoetwaterkreek

Vlakte van Zwijndrecht

Schor Ouden Doel

Paardenschor

Ekers Moeras

Grote Weel

Fort Lillo

Steenlandpolder

Reigersbos

Meeuwenbroedplaats

Fort Liefkenshoek

Kleine Weel

LuithagenLoghidden city

Berendrecht

Oosterweel

Oorderen

Blauwhoef

Wilmarsdonk

Land Van Waas

Straatsburgbrug

Noordlandbrug

Lillobrug

Noordkasteel-bruggen

Meestoofbrug

Melselebrug

Zandvlietbrug

Farnesebrug

Van Cauwelaertbrug

Oudendijkbrug

Berendrechtbrug

Boudewijnbrug

Kruisschansbrug

Oosterweelbrug

Frederik-Hendrikbrug

Wilmarsdonkbrug

1 9 83

67

999795

S6S7

49

36 34

91

105A

S3

48

93

73

M6M2

24

39

23

S4S5

2A

14

32

M3

4B

S2S1

75

19

S8

55

M9

44

62 6147

12

661

665

164 168158

152154

146142

407409

411230 202

221

212 206224 220228

247

285

203207

232

131

101

107

184

54B

117

104105

102100

103

313

253

277279

249

319

261

29C29E

S14

S15

S16

22B

37C

701

663

118120

122

135

667

482480478 476

334

529

521

706

531 520

705

140

124

130 132 136

526

649

647

643645

223

610

614

611

623617

490496 494 488 484

406404

352350

408

250252

300256

248246

332

326

214

283 281

533

528

524

525

527

269

178

307305303 311309

367369

301

485491 487

493

427

401

M10

M11

477475

425

463449

443

437

467

469

471

459461

457451

453

517

498

502

506

504511509

507505

513

236238

SBASRP

474503

455

275

255

273

421

713769

357

775

771

S23S24

S26S27

S29 16A

243 245241

237

229

235

S28

115

770

772

760762

702

624

768

764766

622

S21

616

712708

756

704

754

734732730

716724726728

720

736738742746750

516

541

540534

395

385

363

315 317

620

608606 650

604

119

127186

121

560

405

234

174172

536

615613

304

SIO

S22

400

508

338340

415

419240

242

550

359

703

190194196

S9B

S11S12

27B

361373371

213219 211209

216 208

399397

SPO

321

355

324318

322

316314

336

700

618

518

522

402

328

465

412

342

244

464430428424422420418416414501

466468470472

348

306

312

308

346

483479

188

530532

723727 761

719763

765715

767

729

707

709

28A28C

43A

SIZ

515

711

757

759

349347353 351

343

669

500

SAOW

SLIL

1267

S17BS18A

S13B

1800

S857

S855

S909S907

S905

S863

S865

S859

S903

S867

S901

S869

1930

1702

1704

19201910

1548

DDM5

627B625C

621A625B

S17A

S861

473A473C

473D

403B

1218

1328

1331

1315

13131329

423C

SPPN

433B

1167

1071

1025

1950

1974

SPCN

267A

1706

1940

1700

1742

1099

10871089

755B

753A

751A

16681359

1664

19331934

SPVD

S853

1744

1622

1618

510C512F

514C514B

512C

DDM2

629B

1007

1005

1145

1053

1054

1998

1602

1793

1949

SPCZ

SLIH

1075

S19A

1099

1993

473F

DDM6

423A

306B

SLIH

SCASSPPZ

S913S915

1269

1355

1219

1127

1199 1520

1999

1181

1193 1103

1253

1307

1353

1301

13491351

1600

1616

1560

1608

1562

1746

1606

1602

16121610

1241

12431245

1604

1726

1724

17201716

1714

1009

1992

1968

1165

1261

1311

1644

1740

1237

1235

1231

1233

12211223

1213

1211

1147

1209

1123

1197

1125

1113

1143

12011203

1140

1100

1556

1568

1069

1067

1512

1938

11661268

1337

13091335

1333

1339

13431341

1345

1650

16541652

1640

1642

16481646

1738

1736

1734

1730

1743

1728

170817321710

17121660

1658

1215

1225

1229

1227

1900

1992

1996

1558

1554

1183

1175

1239

1504

1111 11071109

11211119

1251

16281626

1636

1638

1305

12471249

1303

16301632

1634

1013

1023

1614

1948

1620A

500bis

447bis

200bis

1742bis

758748 744

752

740

710714 718 722

492 486

410 426

320330

344

310

226 222 210218192

182 180 176170

166162160156

150148

138134

126128

231227 225 217

129

113

251 257

271

417

413

S911

239

BASF

ICO

3M

COVESTRO

EVONIK

AET

ELECTRABEL

INEOS OXIDE

TOTALRAFFINADERIJ

EXXONMOBIL

A.B.T.

LANXESS

MONSANTOEUROPE

NOVA NATIE

ADPO

ATPC

LANXESS

AMORAS

INDAVER

ATPC

SOLVAY

ATO

BNFW

SOLVAY

BNFW

ASHLAND

LBC

IKO

KATOEN NATIETERMINALS

MSC PSAEUROPEAN TERMINAL

(MPET), K1742

BOREALISKALLO

GUNVORPETROLEUM ANTWERPEN

PSA BREAKBULK

INDAVERSLECO

ANTWERP GATEWAYTERMINAL

EXXONMOBILCHEMICAL

A.B.T.

ARLANXEO BELGIUM

OILTANKINGSTOLTHAVEN

A.B.T.

ANTWERP CONTAINERTERMINAL

EXXONMOBIL

PSA ANTWERPEUROPA TERMINAL

MEXICONATIE

INEOSPHENOL

DREDGING INT.

MEDREPAIR S11 LUBRIZOL

VESTATERMINAL

BNFW

ENGINEDOCK REPAIR

LAWTER

BOREALIS

TOTAL OLEFINS ANTWERP

SEA TANKTERMINAL, K700B

SHIPIT

TOTAL OLEFINS ANTWERP

MAIN HUB(IFB)

W.W.R.S.

KURARAY

BNFW

INOVYN MANUFACTURINGBELGIUM

EUROPORTS TERMINALSLEFTBANK

ANTRECO

MONUMENTCHEMICAL

A.B.T.

ALCA

BP/BLP

SEA-TANKTERMINAL, K510

SEA-TANKTERMINAL

CARGILL

INEOS OLEFINS &POLYMERS EUROPE

STRAATSBURGDOK

VOPAK TERMINALLEFTBANK

SALLAUMTERMINAL

IBOGEM

VOPAK TERMINALEUROTANK

B.S.T.

CNHTRACTOR

EASTMAN

SEA-TANKTERMINAL, K300

EUROPORTS TERMINALSANTWERP

HAZOP

HANDICOTERMINALS

5DEHAVENDOK

MAC² SOLUTIONS

CNHIBELGIUM

ITC RUBISTERMINAL ANTWERP

TOTAL POLYMERSANTWERP

ARKEMA

ORGACOM

SAINT-GOBAINCONSTRUCTION PRODUCTS

FERTIKAL

TOTAL OLEFINSANTWERP

PSA ANTWERPCHURCHILL TERMINAL

LUMIPAPER

WERF & VLAS NATIE

VOLLERSBELGIUM

ZUIDNATIE

SPECIALTY POLYMERSANTWERP

DISTRI ZUID

SEA-TANKTERMINAL

COILTERMINAL

NIPPONSHOKUBAI

PRAXAIR

ZUIDNATIE BREAKBULKINDAVER

DOW

ESSENT

ANTWERP STONETERMINAL (AST)

VOPAK TERMINALACS

HUPAC INTERMODAL

TOTAL RAFFINADERIJANTWERPEN

EUROPORTS TERMINALSANTWERP DIV. MANUFERT

PSA ANTWERPNOORDZEE TERMINAL

INDUSS I

ENGIE FABRICOM

GRC KALLO

NOVAEDES

5DEHAVENDOK

VOTORANTIMTERMINAL

OILTANKING ANTWERPGAS TERMINAL

BILFINGER ROB

MESSERBELGIUM

KATOEN NATIE

KATOEN NATIE TERMINALS

EUROPORTSCONTAINERS, K524

GRENS INSPECTIEPOST (GIP)

KUWAITPETROLEUM

SCR SIBELCO

SYNEGIS

MAINTENANCEPARTNERS

BECO2

VANHEEDEENVIRONMENTAL LOGISTICS

SIKABYK

NOORD NATIETERMINALS

WIJNGAARD NATIETERMINALS

C.STEINWEGBELGIUM

HOLCIM CEMENTANTWERPEN

DEMAKOMENNENSBELGIUM

MOMENTIVE

EUROFRUITPORTS

SAGREXKALLO

EUROCHEM

COMBINANT

LBCCEPSA

AIR LIQUIDE INDUSTRIES BELGIUM

INEOSSTYROLUTION

FRXPOLYMERS

ANTWERPCOLD STORES

NEDERLAND

kerncentrale Doel

LuchthavenDeurne

Vormingsstation Antwerpen-Noord

CentraalStation

Blue Gate Antwerp

BEVEREN

STABROEK

KAPELLEN

ANTWERPEN

BRASSCHAAT

ZWIJNDRECHT

DOEL

LILLO

KALLO

PUTTE

EKEREN

BURCHT

DEURNE

VRASENE

MERKSEM

MELSELE

HOBOKEN

BERCHEM

HOEVENEN

ZANDVLIET

BORGERHOUT

KIELDRECHT

VERREBROEK

LINKEROEVER

BERENDRECHT

PROSPERPOLDER

±Antwerp - Rhine Port

PARISF R A N C E

SCHELD

T

G E R M A N Y

RHINE

N E T H E R L A N D S

Amsterdam

RHINE

RHINE

Ludwigshafen

B E L G I U M

BRUSSELS

LUX.

Gent

Zeebrugge

Rotterdam

Antwerpen

NORTH SEA

Köln

Liège

Basel

Mainz

NeussVenlo

MoerdijkNijmegen

Frankfurt

StuttgartStrasbourg

Germersheim

Ottmarsheim

Valenciennes

Weil am Rhein

Halluin

Duisburg

Gorinchem

© 2018 Port of Antwerp

Uitgave: September 2018

0 1.000 2.000500m

Distributiezone / Distribution area

Watergebonden goederenbehandeling / Quayside cargo handling

Bestaande industrie / Existing industry

Beschikbaar haventerrein / Available port site

Bestaande waterweg / Existing waterway

Woonkern / Residential area

Groene ruimte / Green area

wvb - wegverbinding

spoor

Werken in uitvoering / Work in progress

Bron: havenGIS


Figure 2: Antwerp inland connectivity

4. Methodology In order to improve the different issues that arise related to the port, the terminal and the urban area, our main goal has been to define a solution that allows to obtain an improved multimodal split of the containers that arrive at the terminal and are needed to be sent to final destination. The idea is to promote data sharing between the different key stakeholders in order to obtain the data required to create a tool to be able to develop a predictive and prescriptive model based on real and historical data, that allows the different stakeholders to choose the best option to optimize the supply chain. To achieve the goal, we are developing the CFO, a tool that allows to improve the internal organization of the terminals as well as gives to the carrier different alternatives in order to decide the best choice related to inland mode of transportation. The CFO in focused on the following goals:

a) To improve organization of containers placed on port’s terminal Development a prediction model based on real data (both historical and real time) provided by the different key stakeholders in order to obtain in advance the terminal occupancy as well as the inland mode of transportation expected to be used.


Figure 3: Improve organization of containers model

To improve the organization of containers, data sharing will be the base to create a two-phases predictive

model where first a prediction based on historical data is made and – in real-time and constantly updated –

a second prediction optimizes the final output adjusting the initial results obtained.

b) To predict the best available inland transportation to achieve final destination Analyse the inland requirements of the container as well as the inland transportation connections in order to predict the best existing available mode of transport

Figure 4: Inland transportation mode-Availability model

To predict the availability of inland transportation mode, a two-step scheme is created. First of all, a modelling of the different options is created. Following, a simulation based on the demand expected allows to recommend the best option based on the availability of the different existing modes of transport.

c) To propose new shared services related to inland transportation Analyse the different destinations related to containers in order to propose the creation of on-demand inland routes in order to optimize this inland transportation

Datasets

Predictive model

Improve organization

• Port Authority

• Terminal operator

• Inland transport operators

Output• Terminal occupancy

• Inland mode of transport expected

Data sharing

Historical

Real Time a) Creation of a 1st

output based on historical data

b) Optimization of the 1st output incorporating real time data

Data Lake

Big Data Big&Small Data

Datasets

• Terminal occupancy


Inland transportation mode - Availability

• Inland transport connections

• Capacity of transport connections

Out

put

Inla

nd

Data collection

Real Time

Predictive

Predictive model

Modelling Simulation Recommendation

Machine Learning

Output• Prediction availability of inland

transport routes based on transport time and cost of the route


Figure 5: Inland transportation mode-New Shared services model

To prescript new shared services, a two-step scheme similar to the previous goal described is developed. First, a modelling of the different options is created. Following, a simulation based on the demand expected allows to propose new shared services combining different modes of transport.

5. Next steps and conclusions The project began in May 2018 and it is expected to be finalized in April 2021.

Figure 6: Cargo Flow Optimizer Plan

Currently, the different scenarios as well as the data required to develop the CFO have been defined and we are just receiving the data in order to develop a first version of the CFO to be submitted in August 2019. Acknowledgement This work is a part of the COREALIS project. COREALIS has received funding from the European Union’s Horizon 2020 research & innovation programme under grant agreement no 768994. Content reflects only the authors’ view and European Commission is not responsible for any use that may be made of the information it contains.

Datasets

• Terminal occupancy


Inland transportation mode – Proposal of new shared services

• Inland transport connections

• Capacity of transport connections

Out

put

In

land

Data collection

Real Time

Predictive

Prescriptive model

Modelling Simulation Prescription of new shared services

Machine Learning

Output• New shared transportation services

proposal

• Prediction availability of inland transport routes based on transport time and cost of the routeO

utp

ut 2


References

1. Notteboom, T. (2018). PortGraphic: the top 15 container ports in Europe in 2017. Retrieved January 3, 2019 from https://www.porteconomics.eu/2018/02/28/portgraphic-the-top-15- container-ports-in-europe-in-2017/

2. Corealis website, https://www.corealis.eu

3. https://www.porttechnology.org/news/the_worlds_top_30_container_ports

port multimodal inland mode of transportation predictor ... - …€¦ · the proposed tool, aims...

Documents