scroll identifier for mobile
main-content

## Über dieses Buch

This book constitutes the refereed proceedings of the 7th International Conference on Computational Logistics, ICCL 2016, held in Lisbon, Portugal, in September 2016.

The 29 papers presented in this volume were carefully reviewed and selected for inclusion in the book. They are organized in topical sections entitled: container terminals and maritime transportation; intermodal transport; location and routing; (general) logistics and supply chain management.

## Inhaltsverzeichnis

### A Multi-product Maritime Inventory Routing Problem with Undedicated Compartments

Abstract
This paper considers the problem of routing bulk tankers to minimize cost while managing the inventory in ports. Multiple non-mixable products are transported and the allocation of products to undedicated compartments onboard the ships is an important aspect of the problem. A mixed integer programming formulation of the problem is proposed, and the model is strengthened by including several valid inequalities. Computational results are reported for an evaluation of the model and the valid inequalities. Results are also reported for two simplified models where either the compartments are dedicated or the products are mixable.
Elise Foss, Trine N. Myklebust, Henrik Andersson, Marielle Christiansen

### A MIP Based Local Search Heuristic for a Stochastic Maritime Inventory Routing Problem

Abstract
We consider a single product maritime inventory routing problem in which the production and consumption rates are constant over the planning horizon. The problem involves a heterogeneous fleet of ships and multiple production and consumption ports with limited storage capacity. In spite of being one of the most common ways to transport goods, maritime transportation is characterized by high levels of uncertainty. The principal source of uncertainty is the weather conditions, since they have a great influence on sailing times. The travel time between any pair of ports is assumed to be random and to follow a log-logistic distribution. To deal with random sailing times we propose a two-stage stochastic programming problem with recourse. The routing, the order in which the ports are visited, as well as the quantities to load and unload are fixed before the uncertainty is revealed, while the time of the visit to ports and the inventory levels can be adjusted to the scenario. To solve the problem, a MIP based local search heuristic is developed. This new approach is compared with a decomposition algorithm in a computational study.
Agostinho Agra, Marielle Christiansen, Lars Magnus Hvattum, Filipe Rodrigues

### 2D-Packing with an Application to Stowage in Roll-On Roll-Off Liner Shipping

Abstract
Jone R. Hansen, Ivar Hukkelberg, Kjetil Fagerholt, Magnus Stålhane, Jørgen G. Rakke

### A Vessel Pickup and Delivery Problem from the Disruption Management in Offshore Supply Vessel Operations

Abstract
This paper considers a vessel pickup and delivery problem that arises in the case of disruptions in the supply vessel logistics in the offshore oil and gas industry. The problem can be modelled as a multi-vehicle pickup and delivery problem where delivery orders are transported by supply vessels from an onshore supply base (depot) to a set of offshore oil and gas installations, while pickup orders are to be transported from the installations back to the supply base (i.e. backload). We present both an arc-flow and a path-flow formulation for the problem. For the path-flow formulation we also propose an efficient dynamic programming algorithm for generating the paths, which represent feasible vessel voyages. It is shown through a computational study on various realistic test instances provided by a major oil and gas company that the path-flow model is superior with respect to computational performance.
Nils Albjerk, Teodor Danielsen, Stian Krey, Magnus Stålhane, Kjetil Fagerholt

### Path Planning for Autonomous Inland Vessels Using A*BG

Abstract
To meet the transportation demand and maintain sustainable development, many countries are aiming to promote the competitive position of inland waterway shipping in the transport system. Autonomous transport is seen as a possibility for maritime transport to meet today’s and tomorrow’s challenges. In realizing autonomous navigation, path planning plays an important role. Being the most widely used path planning algorithm for robotics and land-based vehicles, in this paper we analyze A* and its extensions for waterborne applications. We hereby exploit the fact that for vessels optimal paths typically have heading changes only at the corners of obstacles to propose a more efficient modified A* algorithm, A*BG, for autonomous inland vessels. Two locations where ship accidents frequently occur are considered in simulation experiments, in which the performance of A*, A*PS, Theta* and A*BG are compared.
Linying Chen, Rudy R. Negenborn, Gabriel Lodewijks

### Agent-Based Support for Container Terminals to Make Appointments with Barges

Abstract
We consider a container terminal that has to make appointments with barges dynamically with only limited knowledge about future arriving barges, and in the view of uncertainty and disturbances. We study this problem using a case study at the Port of Rotterdam, considering a proposed multi-agent system for aligning barge rotations and terminal quay schedules. We take the perspective of a single terminal participating in this system and focus on the decision making capabilities of its intelligent agent. Using simulation, with input settings based on characteristics of the larger terminals within the Port of Rotterdam, we analyze the benefits of our approach. We conclude that a terminal can increase its utilization significantly by using various sources of flexibility in the operational planning.
Martijn Mes, Albert Douma

### A Logic-Based Benders Decomposition Approach to Improve Coordination of Inland Vessels for Inter-Terminal Transport

Abstract
Large seaports usually contain multiple terminals serving container vessels, railways, trucks and other modes of hinterland transportation. Every time an inland vessel enters a seaport, it visits several terminals for loading and unloading containers. A vessel rotation is the sequence in which a vessel visits the different terminals in a large seaport. Currently, in a seaport like the port of Rotterdam, around 40 % of the inland vessels have to spend a longer time in the port area than originally planned, due to the low utilization of terminal quay resources and uncertainty of waiting times at different terminals. To better utilize the terminal resources in the ports, as well as to reduce the amount of time inland vessels spend in the port area, this paper first proposes a new model in which inland vessels coordinate with each other with respect to the arrival, departure time and the number of inter-terminal containers carried, besides their conventional hinterland containers, with the aim to prevent possible conflicts of their rotations. Then, a logic-based Benders’ decomposition approach is proposed to minimize the total time the inland vessels spent in the port. We compare the performance of the proposed approach with the performance of a centralized approach on the aspects of the runtime, solution quality, and three logistical performance indicators. Simulation results show that the proposed approach generates both faster optimal and faster high-quality solutions than the centralized approach in both small and large problem instance.
Shijie Li, Rudy R. Negenborn, Gabriel Lodewijks

### Scenarios for Collaborative Planning of Inter-Terminal Transportation

Abstract
The immense growth of containerized transport and the increasing frequency of calls of mega-vessels at terminals, serving as transshipment points, require powerful planning methods for the efficient fulfillment of inter-terminal transportation tasks. Collaborative planning, and in particular the exchange of tasks among carriers, is a promising instrument for increasing the efficiency of inter-terminal transportation. The exchange of tasks can be organized by auctions performed by the carriers. Three different collaborative planning scenarios are presented in this paper. These scenarios are evaluated by computational experiments. Based on the preferences of terminal operators and the outcome of computational experiments, recommendations for collaborative inter-terminal transportation are derived.
Herbert Kopfer, Dong-Won Jang, Benedikt Vornhusen

### Solving the Robust Container Pre-Marshalling Problem

Abstract
Container terminals across the world sort the containers in the stacks in their yard in a process called pre-marshalling to ensure their efficient retrieval for onward transport. The container pre-marshalling problem (CPMP) has mainly been considered from a deterministic perspective, with containers being assigned an exact exit time from the yard. However, exact exit times are rarely known, and most containers can at best be assigned a time interval in which they are expected to leave. We propose a method for solving the robust CPMP (RCPMP) to optimality that computes a relaxation of the robust problem and leverages this within a solution procedure for the deterministic CPMP. Our method outperforms the state-of-the-art approach on a dataset of 900 RCPMP instances, finding solutions in many cases in under a second.
Kevin Tierney, Stefan Voß

### A Cooperative Approach to Dispatching and Scheduling Twin-Yard Cranes in Container Terminals

Abstract
To increase the productivity of the storage yard of a container terminal, two identical yard cranes are often deployed in a yard block. In theory, the productivity of a yard block may be doubled with twin-cranes. However, crane interference may severely lower the combined productivity of the twin-cranes. In this paper, we propose an online job dispatching method for twin yard cranes when side loading is used. The method adopts the non-zero-sum game approach to induce the cooperative behaviour in the dispatching and scheduling of jobs for the two cranes to minimize the total job completion time. A one-step lookahead algorithm and a two-step lookahead algorithm are proposed. We evaluate our algorithms against Ng’s lower bound of total completion time for twin-cranes and against the greedy heuristic Smallest Completion Time-First. Our experiments showed that our dispatching and scheduling algorithm performs very well.
Shell Ying Huang, Ya Li

### Online and Offline Container Purchasing and Repositioning Problem

Abstract
We study the management of containers in a logistic chain between a supplier and a manufacturer in a ramp-up scenario where the demand is stochastic and expected to increase. This paper extends our previous study with deterministic demand. We consider a periodic review system with T periods of R time steps. The supplier sends full containers at every step and receives empty containers every period. We consider positive lead times. To face demand increase, the manufacturer can purchase reusable containers at a setup cost while the supplier can buy single-use disposables. Using a dynamic programming framework, we develop an online exact algorithm and an offline heuristic.
Neil Jami, Michael Schröder, Karl-Heinz Küfer

### Towards Real-Time Automated Stowage Planning - Optimizing Constraint Test Ordering

Abstract
Container stowage planning is a complex task in which multiple objectives have to be optimized while ensuring that the stowage rules as well as the safety and balance requirements are observed. Most algorithms for solving the problem are comprised of 2 parts: a container-location selection mechanism and a constraint evaluation engine. The former selects one or more container-location pairs for allocation iteratively and the latter evaluates whether the selected container-location pairs violate any of the constraints. We observe that, using the same selection mechanism, the order in which the constraints are evaluated can have significant impact on the overall efficiency. We propose Sequential Sample Model (SSM) as an improvement over the existing Random Sample Model (RSM) for analysis of the problem. We present and evaluate several strategies in optimizing the constraint evaluation engine. We show how to achieve the optimal constraint ordering with respect to SSM. However, such ordering requires perfect information on the constraint tests which is impractical. We present alternative strategies and show empirically that their efficiencies are close to the optimum. Experiments show that, when compared to an arbitrary ordering, an average of 2.42 times speed up in the evaluation engine can be achieved.
Zhuo Qi Lee, Rui Fan, Wen-Jing Hsu

### Optimizing Train Load Planning: Review and Decision Support for Train Planners

Abstract
Train load planners are confronted with complex practical considerations during the booking and planning process. In order to optimally utilize the available loading space, train capacity is monitored in terms of available length and weight while accounting for the urgency with which load units must be sent. Furthermore, the execution of the load plan by the terminal operator must be performed efficiently to minimize total handling costs. The contribution of this paper is threefold. First, current literature on train load planning is reviewed based on three main groups of factors influencing the load plan composition. Second, a static model is developed to introduce a number of practical constraints from the viewpoint of the network operator. Finally, the model is adapted to reflect the planning environment of a real-life case study.
Hilde Heggen, Kris Braekers, An Caris

### Analysis of Cost Allocation Techniques for Freight Bundling Networks in Intermodal Transport

Abstract
In order to improve the competitive position and efficiency level of intermodal transport, consolidation of freight flows is often suggested. Bundling networks require cooperation between multiple partners in the intermodal transport chain. In this context, the question rises how benefits may be allocated fairly among the participants in the cooperation. A great deal of scientific literature reports on the behavior of allocation methods in collaborations between shippers or carriers making use of unimodal road transport. However, research on cost or savings allocation methods in intermodal transport is scarce. Moreover, since various types of vessels with differing price structures may be considered in intermodal barge transport, the application of allocation mechanisms is not so straightforward compared to a unimodal environment. The main contribution of this paper is thus to provide a first insight in the complexity of sharing cost savings fairly amongst shippers who bundle freight flows in order to reach economies of scale in intermodal barge transport. By applying three different allocation methods, a comparison is made between simple and straightforward allocation mechanisms and more advanced techniques based on cooperative game theory. Special attention is also paid to the stability of the found solutions. The situation of three-, four- and five-partner coalitions is investigated, both for partners with an equal and an unequal amount of shipments. For these six situations, the case of a common barge trajectory and a common end terminal are studied.
Katrien Ramaekers, Lotte Verdonck, An Caris, Dries Meers, Cathy Macharis

### Service and Transfer Selection for Freights in a Synchromodal Network

Abstract
We study the problem of selecting services and transfers in a synchromodal network to transport freights with different characteristics, over a multi-period horizon. The evolution of the network over time is determined by the decisions made, the schedule of the services, and the new freights that arrive each period. Although freights become known gradually over time, the planner has probabilistic knowledge about their arrival. Using this knowledge, the planner balances current and future costs at each period, with the objective of minimizing the expected costs over the entire horizon. To model this stochastic finite horizon optimization problem, we propose a Markov Decision Process (MDP) model. To overcome the computational complexity of solving the MDP, we propose a heuristic approach based on approximate dynamic programming. Using different problem settings, we show that our look-ahead approach has significant benefits compared to a benchmark heuristic.
Arturo Pérez Rivera, Martijn Mes

### A Revenue Management Approach for Network Capacity Allocation of an Intermodal Barge Transportation System

Abstract
We propose a revenue management (RM) model for the network capacity allocation problem of an intermodal barge transportation system. Accept/reject decisions are made based on a probabilistic mixed integer optimization model maximizing the expected revenue of the carrier over a given time horizon. Probability distribution functions are used to characterize future potential demands. The simulated booking system solves, using a commercial software, the capacity allocation problem for each new transportation request. A conventional model for dynamic capacity allocation considering only the available network capacity and the delivery time constraints is used as alternative when analyzing the results of the proposed model.
Yunfei Wang, Ioana C. Bilegan, Teodor Gabriel Crainic, Abdelhakim Artiba

### LORE, A Decision Support Tool for Location, Routing and Location-Routing Problems

Abstract
LOcation Routing Exploration (LORE) is a decision support tool for addressing location, routing and location-routing problems. In this paper the LORE tool will be presented, and its main characteristics addressed. Among the main features of the tool is the ability to support a variety of problems currently being studied in the location and routing literature (due to the proposed data structure), and the graphical user interface (GUI). The data structure will be presented being provided an explanation on how it can support related problems. The GUI main goal is not only to aid the solution-finding process but also to foster greater insight into the problem(s) at hand. To that extent, the GUI, developed to fit the target user’s profile and intended tasks, is presented, namely data input and visualization features.
Rui Borges Lopes, Carlos Ferreira, Beatriz Sousa Santos

### Two Echelon Location Routing Problem with Simultaneous Pickup and Delivery: Mixed Integer Programming Formulations and Comparative Analysis

Abstract
This paper addresses the two-echelon location routing problem with simultaneous pickup and delivery (2E-LRPSPD). The 2E-LRPSPD deals with optimally locating primary and secondary facilities, and integrating goods distribution from depots and collection from customers and secondary depots. To the best of our knowledge there is no previous study on this problem. We propose two mixed integer programming formulations for the 2E-LRPSPD. While the first formulation is a two-index node-based formulation, the second one is a two-index flow-based formulation. Moreover, a family of valid inequalities are adapted from the literature to strengthen the formulations. In order to evaluate the performances of the formulations and valid inequalities, we conduct an experimental study on the instances derived from the literature. The computational results show that the flow-based formulation produces better lower bounds than the node-based formulation on small and medium-size problems.
Ece Arzu Demircan-Yildiz, Ismail Karaoglan, Fulya Altiparmak

### Vehicle Routing for Fleets with Electric- and Combustion-Powered Vehicles

Abstract
Optimal transportation plans for fleets with electric-powered vehicles (EPVs) differ substantially from plans generated for fleets with combustion-powered vehicles (CPVs). The main reasons for this difference are the reduced range and payload of EPVs (compared to CPVs) as well as their increased efficiency. In this paper, transportation plans for CPVs and EPVs which must not be recharged during route fulfillment are analyzed by computational experiments. The advantages of CPVs with respect to totally driven distances, number of used vehicles and the ability to generate feasible plans are opposed to the advantages of EPVs with respect to $$CO_2$$ emissions. Additionally it is shown that the specific drawbacks of CPVs and EPVs can be mitigated by exploiting the flexibility of a fleet which is composed of both, EPVs and CPVs.
Herbert Kopfer, Kristian Schopka

### The Bi-Objective k-Dissimilar Vehicle Routing Problem

Abstract
This paper deals with the k-dissimilar vehicle routing problem in which a set of k dissimilar alternatives for a Capacitated Vehicle Routing Problem (CVRP) has to be determined for a single instance. The tradeoff between minimizing the longest routing alternative and maximizing the minimum dissimilarity between two routing alternatives is investigated. Since short vehicle routings tend to be similar to each other, a conflict of objectives arises. The developed heuristic approach approximates the Pareto set with respect to this tradeoff using a dissimilarity metric based on a grid. The method is tested on benchmark instances of the CVRP and findings are reported.
Sandra Zajac

### A Branch-and-Price Algorithm for the Vehicle Routing Problem with 2-Dimensional Loading Constraints

Abstract
In this paper, we describe a branch-and-price algorithm for the capacitated vehicle routing problem with 2-dimensional loading constraints and a virtually unlimited number of vehicles. The column generation subproblem is solved heuristically through variable neighborhood search. Branch-and-price is used when it is not possible to add more attractive columns to the current restricted master problem, and the solution remains fractional. In order to accelerate the convergence of the algorithm, a family of valid dual inequalities is presented. Computational results are provided to evaluate the performance of the algorithm and to compare the different branching strategies proposed.
Telmo Pinto, Cláudio Alves, José Valério de Carvalho

### The Static Bicycle Repositioning Problem - Literature Survey and New Formulation

Abstract
This paper considers the static bicycle repositioning problem (SBRP), which deals with optimally re-balancing bike sharing systems (BSS) overnight, i.e. using service vehicles to move bikes from (nearly) full stations to (nearly) empty stations. An exhaustive literature survey comparing existing models is presented, and a new and improved mathematical formulation for the SBRP is proposed. The model is tested on a number of instances generated based on data from a real BSS.
Hans Martin Espegren, Johannes Kristianslund, Henrik Andersson, Kjetil Fagerholt

### Service Network Design of Bike Sharing Systems with Resource Constraints

Abstract
Station-based bike sharing systems provide an inexpensive and flexible supplement to public transportation systems. However, due to spatial and temporal demand variation, stations tend to run full or empty over the course of a day. In order to establish a high service level, that is, a high percentage of users being able to perform their desired trips, it is therefore necessary to redistribute bikes among stations to ensure suitable time-of-day fill levels. As available resources are scarce, the tactical planning level aims to determine efficient master tours periodically executed by redistribution vehicles. We present a service network design formulation for the bike sharing redistribution problem taking into account trip-based user demand and explicitly considering service times for bike pick-up and delivery. We solve the problem using a two-stage MILP-based heuristic and present computational results for small real-world instances. In addition, we evaluate the performance of the master tours for multiple demand scenarios.
Bruno Albert Neumann-Saavedra, Teodor Gabriel Crainic, Bernard Gendron, Dirk Christian Mattfeld, Michael Römer

### An Agent-Based Simulation Framework to Evaluate Urban Logistics Schemes

Abstract
Inefficient urban freight transport has a negative impact on both livability in cities and profit margins in the supply chain. Urban logistics schemes, consisting of governmental policies and company initiatives, attempt to address these problems. However, successful schemes are difficult to realize due to the divergent objectives of the agents involved in urban logistics. Traditional optimization techniques fall short when evaluating schemes, as they do not capture the required change in behavior of autonomous agents. To properly evaluate schemes, we develop an agent-based simulation framework that assesses the interaction between five types of autonomous agents. Compared to existing studies in this field, we contribute by (i) explicitly including company-driven initiatives, and (ii) adopting a supply chain-wide perspective. We illustrate the working of our framework by testing a number of schemes on a virtual network.
Wouter van Heeswijk, Martijn Mes, Marco Schutten

### Continuous-Time Formulation for Oil Products Transportation Scheduling

Abstract
This paper presents a novel Mixed Integer Linear Programming (MILP) model for the operational planning of an oil transportation system characterized by a straight multiproduct pipeline with dual purpose terminals that could represent facility input or output. It is based on a continuous representation in both time and volume scales and is capable of meeting all operational constraints related to product sequencing, mass balances and pipeline loading/unloading operations. Contrary to previous approaches, the model allows an intermediate node and the previous segment to simultaneously inject material in the pipeline. Computational results and data are reported.
Hossein Mostafaei, Pedro M. Castro

### Impact of Collaborative Decision Making in Optimized Air Traffic Control: A Game Theoretical Approach

Abstract
Air traffic is growing, putting increasing stress to airports and air traffic control. The introduction of optimized approaches, based on mathematical optimization paradigms for planning and real time control, can be a possible solution to this issues. We investigate the practical setting of an advanced optimization algorithm in a real-life setting of a major airport where traffic is diverse, belonging to multiple companies. We compare to the incumbent practice (based on First Come First Served) in order to determine a gap with optimized solutions computed by advanced algorithms. Those are based on a job shop scheduling model and solved by a commercial solver.
This paper analyses the benefit for the involved operators of such approaches by associating a monetary cost/benefit to operations. Cooperative game theory tools have been used in the analysis. In particular, we use the Shapley value to determine the fair distribution of the costs based on the marginal improvement that the optimization of the traffic belonging to any airline brought to the system. The main conclusions of this study are the determination of the superior performance in terms of minimising the delay experienced by the whole airport, which reaches more than 25 %. The benefit allocation gives share of benefits more insightful than a simple proportional approaches based on share of traffic, or share of delay. The practical implications of the analysis with regard to variety in benefits as well as possible implementations by the different operators and companies are also analysed.
Manish Tripathy, Marcella Samà, Francesco Corman, Gabriel Lodewijks

### Impact of Dwell Time on Vertical Transportation Through Discrete Simulation in SIMIO

Abstract
This work has the objective of simulating an elevator system, using SIMIO software. Firstly, two different approaches, and its implementation, will be explained and compared: Vehicle vs. Entity. After selecting the Entity-approach, due to its more flexible processes and the limitations of the Vehicle-approach, it will be used to conduct the simulation experiments. The purpose is to evaluate the impact of dwell time - time in which the elevator remains stopped, allowing for clients to enter and exit - in the performance of the system. That will be achieved analysing the impact on the total time - spent by clients from placing a call until reaching its destination - number of clients inside the system and waiting for the elevator, waiting time, elevator occupation and number of elevator movements. The analysis of the results indicates that, for the properties defined, the best time for the elevator to stay with its doors open is around 10 seconds.
Marcelo Henriques, António A. C. Vieira, Luís M. S. Dias, Guilherme A. B. Pereira, José A. Oliveira

### Improving Order Picking Efficiency by Analyzing Combinations of Storage, Batching, Zoning, and Routing Policies

Abstract
In order to differentiate from competitors in terms of customer service, warehouses accept late orders while providing delivery in a quick and timely way. This trend leads to a reduced time to pick an order. The objective of this research is to simulate and evaluate the interaction between several storage, batching, zone picking and routing policies in order to reduce the order picker travel distance. The value of integrating these four operation policy decisions is proven by a real-life case study. A full factorial ANOVA provides insight into the interactions between storage, batching, zoning, and routing policies. The results of the study clearly indicate that warehouses can achieve significant benefits by considering storage, batching, zone picking, and routing policies simultaneously. Awareness of the influence of an individual policy decision on the overall warehouse performance is required to manage warehouse operations, resulting in enhanced customer service.
Teun van Gils, Kris Braekers, Katrien Ramaekers, Benoît Depaire, An Caris

### Improving Production Logistics Through Materials Flow Control and Lot Splitting

Abstract
Competitive advantage of make-to-order manufacturing companies is highly dependent on their capability to offer short delivery times and on time delivery. This calls for effective production and materials flow control – a core part of production logistics. This paper applies discrete simulation to study the delivery performance of a make-to-order manufacturing system configured as a general flow shop, when operated under two card-based material flow control mechanisms: CONWIP and GKS. The influence of two lot splitting strategies on the performance of these mechanisms is also evaluated. Results show that GKS clearly outperforms CONWIP and that splitting strategies have a positive impact on the performance of both mechanisms. GKS also showed to be particularly robust to the variation of the number of production authorisation cards used. This, together with the fact that the card-based mechanisms require little data handling and simplify production control, makes GKS attractive for practical application in make-to-order companies.
Catarina Gomes, Andreia Ribeiro, João Freitas, Luís Dias, Guilherme Pereira, António Vieira, Nuno O. Fernandes, Sílvio Carmo-Silva

### Backmatter

Weitere Informationen