Search

This paper presents a logistics serious game that describes an anticipatory planning problem for the dispatching of trucks, barges, and trains, considering uncertainty in future container arrivals. The problem setting is conceptually easy to grasp, yet difficult to solve optimally. For this problem, we deploy a variety of benchmark algorithms, including two heuristics and two reinforcement learning implementations. We use the serious game to compare the manual performance of human decision makers with those algorithms. Furthermore, the game allows humans to create their own automated planning rules, which can also be compared with the implemented algorithms and manual game play. To illustrate the potential use of the game, we report the results of three gaming sessions: with students, with job seekers, and with logistics professionals. The experimental results show that reinforcement learning typically outperforms the human decision makers, but that the top tier of humans come very close to this algorithmic performance.

Logistics operations have a direct impact on the effectiveness of the humanitarian relief operations and the survival of the population, supplying all demands in a short period of time using the available limited resources. This work addresses the Emergency k-Location Routing Problem (EkLRP) where humanitarian and relief aid has to be distributed from medical infrastructure to the affected people by routing emergency-aimed vehicles minimizing the time required to provide the humanitarian aid. This work proposes a Simulated Annealing with temperature reset in order to promote diversification as well as for escaping from local optima. The numerical experiments indicate that the metaheuristic approach proposed to solve the EkLRP reports high-quality solutions in reasonable computational times.

Simulated annealing (SA) is a well-known metaheuristic commonly used to solve a great variety of \(\mathcal {NP}\)-hard problems such as the quadratic assignment problem (QAP). As commonly known, the choice and size of neighborhoods can have a considerable impact on the performance of SA. In this work, we investigate and propose a SA variant that considers variable neighborhood structures driven by the state of the search. In the computational experiments, we assess the contribution of this SA variant in comparison with the state-of-the-art SA for the QAP applied to printed circuit boards and conclude that our approach is able to report better solutions by means of short computational times.

In this paper we tackle the container allocation problem in multimodal transportation planning under uncertainty in container arrival times, using Deep Reinforcement Learning. The proposed approach can take real-time decisions on allocating individual containers to a truck or to trains, while a transportation plan is being executed. We evaluated our method using data that reflect a realistic scenario, designed on the basis of a case study at a logistics company with three different uncertainty levels based on the probability of delays in container arrivals. The experiments show that Deep Reinforcement Learning methods outperform heuristics, a stochastic programming method, and methods that use periodic re-planning, in terms of total transportation costs at all levels of uncertainty, obtaining an average cost difference with the optimal solution within 0.37% and 0.63%.

The Traveling Salesman Problem (TSP) is a well-known optimization problem with a wide range of extensions and applications in delivery systems. In this paper, we consider a recent extension of the TSP where a truck in collaboration with a single drone should visit a set of customers while minimizing the transportation times. We propose a Variable Neighbourhood Search (VNS) and a Multi-Start VNS (MS-VNS) algorithm, develop new neighbourhood structures, and compare the solutions against an existing mixed-integer linear programming (MILP) formulation. We take a set of instances based on existing benchmarks from the related literature. Results point out that the new neighbourhood structures substantially improve the performance of the VNS algorithms. Furthermore, results also show that the exact method is only able to find competitive solutions for small sets of instances, whereas our MS-VNS approach reaches better solution quality for large instances.

The shared autonomous vehicle (SAV) is a new concept that meets the upcoming trends of autonomous driving and changing demands in urban transportation. SAVs can carry passengers and parcels simultaneously, making use of dedicated passenger and parcel modules on board. A fleet of SAVs could partly take over private transport, taxi, and last-mile delivery services. A reduced fleet size compared to conventional transportation modes would lead to less traffic congestion in urban centres. This paper presents a method to estimate the optimal capacity for the passenger and parcel compartments of SAVs. The problem is presented as a vehicle routing problem and is named variable capacity share-a-ride-problem (VCSARP). The model has a MILP formulation and is solved using a commercial solver. It seeks to create the optimal routing schedule between a randomly generated set of pick-up and drop-off requests of passengers and parcels. The objective function aims to minimize the total energy costs of each schedule, which is a trade-off between travelled distance and vehicle capacity. Different scenarios are composed by altering parameters, representing travel demand at different times of the day. The model results show the optimized cost of each simulation along with associated routes and vehicle capacities.

The craft beer supply chain in the USA differs from the supply chain of macro breweries in its structure, handled volumes and product shelf-life. In this work, we study how these smaller craft breweries can benefit from transparency in their supply chain. We consider additional information sharing of orders and inventories at downstream nodes. The levels that we investigate grant the brewery incremental access to distributor, wholesaler, and retailer data. We show how this knowledge can be incorporated effectively into the brewery’s production planning strategy. Extending the well-known beer game, we conduct a simulation study using real-world craft beer supply chain parameters and demand. We quantify the impact of information sharing on the craft brewery’s sales, spoilage, and beer quality. Our model is designed to directly support the brewery when evaluating the value of downstream information and negotiating data purchases with brokers. Through a computational analysis, we show that the brewery’s benefits increase almost linearly with every downstream node that it gets data from. Full transparency allows to halve the missed beer sales, and beer spoilage can even be reduced by 70% on average.

In this work, the Multi-Depot Green VRP with Pickups and Deliveries (MDGVRP-PD) is studied. It is a routing optimization problem in which the objective is to construct a set of vehicle routes considering multiple depots and one-to-one pickup and delivery operations that minimize emissions through fuel consumption, which depends on weight and travel distance. In one-to-one problems, goods must be transported between a single origin and its single associated destination. Practical considerations imply addressing the pickup and delivery of customers from multiple depots, where a logistics service company can efficiently combine its resources, thus reducing environmental pollution. To tackle this problem, we develop a mathematical programming formulation and matheuristic approach based on the POPMUSIC (Partial Optimization Metaheuristic under Special Intensification Conditions) framework. The results show that if the weight carried on the routes as part of the fitness measure is considered, our matheuristic approach provide an average percentage improvement in emissions of \(30.79\%\), compared to a fitness measure that only takes into account the distances of the routes.

Smart modular freight containers - as propagated in the Physical Internet paradigm - are equipped with sensors, data storage capability and intelligence that enable them to route themselves from origin to destination without manual intervention or central governance. In this self-organizing setting, containers may autonomously place bids on transport services in a spot market setting. However, for individual containers it might be difficult to learn good bidding policies due to limited observations. By sharing information and costs between one another, smart containers can jointly learn bidding policies, even though simultaneously competing for the same transport capacity. We replicate this behavior by learning stochastic bidding policies in a semi-cooperative multi-agent setting. To this end, we develop a reinforcement learning algorithm based on the policy gradient framework. Numerical experiments show that sharing solely bids and acceptance decisions leads to stable bidding policies. Real-time system information only marginally improves performance; individual job properties suffice to place appropriate bids. Furthermore, we find that carriers may have incentives not to share information with the smart containers. The experiments give rise to several directions for follow-up research, in particular the interaction between smart containers and transport services in self-organizing logistics. Key in this approach is the interplay between the degree of autonomy of logistic systems and their degree of cooperativeness. On these two pillars, a unifying framework is presented, distinguishing four fundamental categories of self-organizing logistics. To illustrate the working of the framework in practice, we present four real-life case studies, one per each category. The case studies are positioned as-is, and concrete directions for (more) self-organization are presented for each case. Moreover, possible additional dimensions of the framework, e.g., control hierarchy, system intelligence, connectivity, and predictability are discussed. The usefulness of the framework established is two-fold: (i) it provides a common ground for researchers to position their work and to identify potential future directions for research and (ii) it serves as a practical and understandable starting point for practitioners on investigating how self-organization may affect their business and where their limited resources should be focused upon.

This paper considers a rich vehicle routing problem in which a combination of transportation costs and customer-perceived waiting times should be minimized and a differentiation is made between priority and non-priority customers. We illustrate the problem using a case study of a wholesaler with its own last-mile delivery network where customers can have pickup and delivery demand and are served by a heterogeneous fleet of vehicles. We propose a bi-objective mathematical problem formulation, minimizing the combination of transportation costs and customer dissatisfaction. We model customer dissatisfaction using a non-linear function that approximates the perceived waiting time of the customers. To be able to solve realistically-sized problems in reasonable time, we propose a Simulated Annealing heuristic, Variable Neighborhood Search, and a combination of these. We perform various experiments considering different customer preferences (visit as soon as possible or at a specific time) and problem settings. For the combined objective, we see an average costs reduction for the dissatisfaction function approach compared to the standard time window approach of 48% over all experiments. Furthermore, we observe an average reduction in perceived waiting time of 48% and 20% for priority and non-priority customers, respectively. Key in this approach is the interplay between the degree of autonomy of logistic systems and their degree of cooperativeness. On these two pillars, a unifying framework is presented, distinguishing four fundamental categories of self-organizing logistics. To illustrate the working of the framework in practice, we present four real-life case studies, one per each category. The case studies are positioned as-is, and concrete directions for (more) self-organization are presented for each case. Moreover, possible additional dimensions of the framework, e.g., control hierarchy, system intelligence, connectivity, and predictability are discussed. The usefulness of the framework established is two-fold: (i) it provides a common ground for researchers to position their work and to identify potential future directions for research and (ii) it serves as a practical and understandable starting point for practitioners on investigating how self-organization may affect their business and where their limited resources should be focused upon.

Inland shipping is a critical part of the global transport network that links population centers over rivers for both cargo and passenger transportation. Inland ships follow tight schedules, leaving little time for required maintenance or emergency repairs. Maintenance providers must therefore send their technicians out to ships during the ships’ operation and quickly perform their work during regular calls on the ships’ schedules. The resulting technician routing problem is challenging for maintenance providers to solve, not only due to time-dependent distance matrix, but also because of variable task durations, technician car pooling and the long planning horizon. We use an extended time space graph to formulate a binary linear model to provide routes for technicians and their vehicles. We evaluate the model on a benchmark of instances inspired from our industrial collaborator, showing that it can find optimal solutions to small and mid-sized instances.

Global intermodal transportation involves the movement of shipments between inland terminals located in different continents by using ships, barges, trains, trucks, or any combination among them through integrated planning at a network level. One of the challenges faced by global operators is the matching of shipment requests with transport services in an integrated global network. The characteristics of the global intermodal shipment matching problem include acceptance and matching decisions, soft time windows, capacitated services, and transshipments between multimodal services. The objective of the problem is to maximize the total profits which consist of revenues, travel costs, transfer costs, storage costs, delay costs, and carbon tax. Travel time uncertainty has significant effects on the feasibility and profitability of matching plans. However, travel time uncertainty has not been considered in global intermodal transport yet leading to significant delays and infeasible transshipments. To fill in this gap, this paper proposes a chance-constrained programming model in which travel times are assumed stochastic. We conduct numerical experiments to validate the performance of the stochastic model in comparison to a deterministic model and a robust model. The experiment results show that the stochastic model outperforms the benchmarks in total profits.

We address a milk run logistics problem in which goods must be transferred between facilities on a regular basis to support a company’s production. The resulting optimization problem is a form of cyclical, multi-commodity inventory routing problem with pick-ups and deliveries. This problem becomes particularly challenging when either all available cargo of a particular commodity must be picked up or none of it, so as to simplify internal logistics procedures. We model the problem mathematically and introduce a two-phase heuristic to solve the problem. The first phase consists of an adaptive large neighborhood search (ALNS), and the second phase uses a relaxed version of the mathematical model to improve the routes for the vehicles. We present experimental results on a dataset based on real-world data from a company in Germany and show that our two-phase procedure can find high quality solutions even to real-world sized problems in reasonable amounts of computation time.

Automated ground vehicles (AGVs) are essential parts of container operations at many ports. Forming platoons—as conceptually established in trucking—may allow these vehicles to directly cater demand points such as dry ports in the hinterland. In this work, we aim to assess such AGV platoons in terms of operational efficiency and costs, considering the case of the Port of Rotterdam. We propose a multi-objective mixed-integer programming model that minimizes dwell and idle times, on the one hand, and the total cost of the system involving transportation, labor, and platoon formation costs, on the other hand. To achieve Pareto optimal solutions that capture the trade-offs between minimizing cost and time, we apply an augmented epsilon constraint method. The results indicate that all the containers are delivered by AGVs. This not only shortens the dwell time of the containers by decreasing loading/unloading processes and eliminating stacking but also leads to considerable cost savings.

When barges are scheduled before the demand for container transport is known, the scheduled departures may match poorly with the realised demands’ due dates and with the truck utilization. Synchromodal transport enables simultaneous planning of container, truck and barge routes at the operational level. Often these decisions are taken by multiple stakeholders who wants cooperation, but are reluctant to share information. We propose a novel co-planning framework, called departure learning, where a barge operator learns what departure times perform better based on indications from the other operator. The framework is suitable for real time implementation and thus handles uncertainties by replanning. Simulated experiment results show that co-planning has a big impact on vehicle utilization and that departure learning is a promising tool for co-planning.

Approximating the solution value of transportation problems has become more relevant in recent years, as these approximations can help to decrease the computational effort required for solving those routing problems. In this paper, we apply several regression methods to predict the total distance of the traveling salesman problem (TSP) and vehicle routing problem (VRP). We show that distance can be estimated fairly accurate using simple regression models and only a limited number of features. We use features found in scientific literature and introduce a new class of spatial features. The model is validated on a dynamic waste collection case in the city of Amsterdam, the Netherlands. We introduce a cost function that combines the travel distance and service level, and show that our model can reduce distances up to 17%, while maintaining the same service level, compared to a well-known heuristic approximation. Furthermore, we show the benefits of using approximations for combining offline learning with online or frequent optimization.

We present a “rich” Petrol Station Replenishment Problem (PSRP) with real-life characteristics that represents the complexities involved in actual operations. The planning is optimised over multiple days and therefore, the new variant can be classified as the Multi-Period Petrol Station Replenishment Problem (MP-PSRP). A Mixed Integer Linear Programming (MILP) formulation is developed and a decomposition heuristic is proposed as a solution algorithm, which is evaluated with a case study from a real-life petrol distributor in Denmark. To determine delivery quantities, the heuristic uses the newly introduced simultaneous dry run inventory policy. A procedure is applied to improve the initial solution. A commercial solver is able to find feasible solutions only for instances with up to 20 stations and 7 days for the MILP model where optimality is guaranteed for instances up to 10 stations and 5 days. The heuristic on the other hand provides feasible solutions for the full case study of 59 stations and 14 days, within a time limit of 2 h.

Port terminals on floating modular platforms are a conceivable solution for the problem of limited space and water depths restrictions of ports in estuary regions. A design of a dedicated Transport&Logistic hub has been developed in the scope of the Horizon 2020 project Space@Sea. This paper addresses dedicated options of waterborne hinterland transports and discusses opportunities for bypassing onshore terminals by means of river-sea or sea-going inland vessels. A tailored simulation method for ship operations utilises a specific cost model and is applied to derived demand scenarios. Cargo flow statistics of an onshore port have been projected onto the hub to identify relevant waterborne transports to the hinterland. Three different vessel types are implemented, whereas inland vessels are considered with two different sizes. A comparison of round trip durations and transport costs per transported container between a floating terminal and a relevant hinterland port pointed out, that a non-stop connection with sea-going inland vessels is the economically favourable solution. A feeder vessel is the faster solution in coastal waters but it can not compensate the time saved by omitted terminal visits on a direct hinterland connection.

Inland waterway transport is becoming attractive due to its minimum environmental impact in comparison with other transportation modes. Fixed timetables and routes are adopted by most barge operators, avoiding the full utilization of the available resources. Therefore a flexible model is adopted to reduce the transportation cost and environmental impacts. This paper regards the route optimization of barges as a pickup and delivery problem (PDP). A Mixed Integer Programming (MIP) model is proposed to formulate the PDP with transshipment of barges, and an Adaptive Large Neighborhood Search (ALNS) is developed to solve the problem efficiently. The approach is evaluated based on a case study in the Rhine Alpine corridor and it is shown that ALNS is able to find good solutions in reasonable computation times. The results show that the cost is lower when there is more flexibility. Moreover, the cost comparison shows that transshipment terminals can reduce the cost for barge companies.

Dynamic pricing can be used for better fleet distribution in free-floating vehicle sharing (FFVS), and thus increase utilization and revenue for the provider by reducing supply-demand asymmetry. Supply-demand asymmetry refers to the existence of an undersupply of vehicles at some locations at the same time as underutilization of vehicles at other locations. We propose to use dynamic pricing as an instrument to incentivize users to rebalance these vehicles from low demand locations to high demand locations. Despite significant research in rebalancing vehicle sharing, the literature so far lacks experimental results on dynamic pricing in free-floating vehicle sharing. We propose to use an algorithm that minimizes the differences in the idle time of vehicles. The algorithm is tested in a real-life experiment that was conducted in cooperation with an FFVS provider. The results of the experiment are not statistically significant, but they clearly indicate that even slight differences in pricing and a simple algorithm can already influence user-behavior to counter supply-demand asymmetry. Improving the existing algorithm with more experimental research is advised to further uncover the potential of this strategy.

Residents of cities’ most disadvantaged areas face significant barriers to key life activities, such as employment, education, and healthcare, due to the lack of mobility options. Shared autonomous vehicles (SAVs) create an opportunity to overcome this problem. By learning user demand patterns, SAV providers can improve regional service levels by applying anticipatory relocation strategies that take into consideration when and where requests are more likely to appear. The nature of transportation demand, however, invariably creates learning biases towards servicing cities’ most affluent and densely populated areas, where alternative mobility choices already abound. As a result, current disadvantaged regions may end up perpetually underserviced, therefore preventing all city residents from enjoying the benefits of autonomous mobility-on-demand (AMoD) systems equally. In this study, we propose an anticipatory rebalancing policy based on an approximate dynamic programming (ADP) formulation that processes historical demand data to estimate value functions of future system states iteratively. We investigate to which extent manipulating cost settings, in terms of subsidies and penalties, can adjust the demand patterns naturally incorporated into value functions to improve service levels of disadvantaged areas. We show for a case study in the city of Rotterdam, The Netherlands, that the proposed method can harness these cost schemes to better cater to users departing from these disadvantaged areas, substantially outperforming myopic and reactive benchmark policies.

The Algorithm Selection Problem seeks to select the most suitable algorithm for a given problem. For solving it, the algorithm selection systems have to face the so-called cold start. It concerns the disadvantage that arises in those cases where the system involved in the selection of the algorithm has not enough information to give an appropriate recommendation. Bearing that in mind, the main goal of this work is two-fold. On the one hand, a novel meta-learning-based approach that allows selecting a suitable algorithm for solving a given logistic problem is proposed. On the other hand, the proposed approach is enabled to work within cold start situations where a tree-structured hierarchy that enables to compare different metric dataset to identify a particular problem or variation is presented.

The Quadratic Knapsack Problem (QKP) is a well-known optimization problem aimed to maximize a quadratic objective function subject to linear capacity constraints. It has several applications in different fields such as telecommunications, graph theory, logistics, hydrology and data allocation, among others. In this paper, we propose the application of a novel population-based metaheuristic referred to as Multi-leader Migrating Birds Optimization (MMBO), which exploits the concepts of cooperation and communication along the search leading to a collective learning, to solve a wide range of well-known QKP instances.

Choosing the correct algorithm to solve a problem still remains an issue 40 years after the Algorithm Selection Problem was first posed. Here we propose a hyper-heuristic which can apply one of two meta-heuristics at the current stage of the search. A scoring function is used to select the most appropriate algorithm based on an estimate of the improvement that might be made by applying each algorithm. We use a differential evolution algorithm and a genetic algorithm as the two meta-heuristics and assess performance on a suite of 18 functions provided by the Generalization-based Contest in Global Optimization (genopt). The experimental evaluation shows that the hybridisation is able to provide an improvement with respect to the results obtained by both the differential evolution scheme and the genetic algorithm when they are executed independently. In addition, the high performance of our hybrid approach allowed two out of the three prizes available at genopt to be obtained.

Maritime container terminals are essential infrastructures in global supply chains. Their high management complexity and heterogeneous processes make them an interesting field to apply heuristics. A brief overview of the main optimization problems found at maritime container terminals and a review of the way they are related to each other are firstly introduced. In order to solve these problems, several heuristics are presented and analyzed. The computational results reveal that they are suitable to be applied in practical scenarios due to the fact that they provide high-quality solutions in short computational times.

The p-Cable Trench Problem is a telecommunications network design problem, which jointly considers cable and trench installation costs and addresses the optimal location of p facilities. In this work, a matheuristic approach based on the POPMUSIC (Partial Optimization Metaheuristic under Special Intensification Conditions) framework is developed. The inspected neighborhoods for building sub-problems include lexicographic as well as nearest neighbor measures. Using benchmark data available from literature it is shown that existing results can be outperformed.

We present novel algorithmic schemes for dealing with large scale continuous problems. They are based on the recently proposed population-based meta-heuristics Migrating Birds Optimisation (mbo) and Multi-leader Migrating Birds Optimisation (mmbo), that have shown to be effective for solving combinatorial problems. The main objective of the current paper is twofold. First, we introduce a novel neighbour generating operator based on Differential Evolution (de) that allows to produce new individuals in the continuous decision space starting from those belonging to the current population. Second, we evaluate the performance of mbo and mmbo by incorporating our novel operator to them. Hence, mbo and mmbo are enabled for solving continuous problems. A set of well-known large scale functions is used for comparison purposes.