Massimo Di Francesco

Maritime repositioning of empty containers under uncertain port disruptions

This paper addresses the problem of repositioning empty containers in maritime networks under possible port disruptions. Since drastically different futures may occur, the decision making process for dealing with this problem cannot ignore the uncertain nature of its parameters. In this paper, we consider the uncertainty of relevant problem data by a stochastic programming approach, in which different scenarios are included in a multi-scenario optimization model and linked by non-anticipativity conditions. Numerical experiments show that the multi-scenario solutions provide a hedge against uncertainty when compared to deterministic decisions and exhibit some forms of robustness, which mitigate the risks of not meeting empty container demand.

Implementing the street-turn strategy by an optimization model

The street-turn option represents a major strategy for the profitability of shipping companies supplying container-based transportation. This option consists in the distribution of trucks delivering loaded containers to import customers, the subsequent allocation of empty containers to export customers and the final dispatch of loaded containers to departure ports. However, the determination of truck routes is a time-consuming activity for shipping companies, because available information can suddenly change while they are making their decisions. In this paper we aim to propose a decision support tool to quickly determine truck routes and implement the street-turn strategy. This tool is based on an optimization model determining the allocation of empty containers between customers and defining truck routes in a post-optimization phase. We compare routes resulting from the proposed model to the decisions of a real shipping company. Early results indicate that this approach represents a promising support for shipping companies in dealing with street-turns. It can significantly reduce distances travelled by trucks and times requested to determine routes.

Regularity diagnosis by Automatic Vehicle Location raw data

Bus regularity is a crucial factor for high frequency public transport systems, because it represents a relevant measure of quality of service for both users and transit agencies. Low regularities for users are associated with bunching phenomena or large gaps between buses, which result in low attractiveness of the service for transit agencies. Therefore, evaluating the regularity is extremely desirable, but may also be a complex task in medium-size cities due to the huge amount of data which must be collected and processed effectively. Automatic Vehicle Location (AVL) technologies, which are particularly used by transit agencies in Western Europe, can address the data collection problem, but they involve several challenges such as correcting anomalies in collected raw data and processing information efficiently. In this paper, we propose a method to automatically handle AVL raw data for measuring the Level of Service (LoS) of bus regularity at each bus stop and time interval of any high frequency route. The results are represented by easy-to-read control dashboards and graphs. We discuss the experimentation of this method in a real case study to provide insights into the detailed characterization of bus regularity. The method is applied to data obtained from the transport agency CTM in Cagliari (Italy), whose vehicles are all equipped with AVL technologies.

An adaptive guidance meta-heuristic for the vehicle routing problem with splits and clustered backhauls

This paper presents the case study of an Italian carrier, Grendi Trasporti Marittimi, which provides freight transportation services by trucks and containers. Its trucks deliver container loads from a port to import customers and collect container loads from export customers to the same port. In this case study, all import customers in a route must be serviced before all export customers, each customer can be visited more than once and containers are never unloaded or reloaded from the truck chassis along any route. We model the problem using an Integer Linear Programming formulation and propose an Adaptive Guidance metaheuristic. Our extensive computational experiments show that the adaptive guidance algorithm is capable of determining good-quality solutions in many instances of practical or potential interest for the carrier within 10 min of computing time, whereas the mathematical formulation often fails to provide the first feasible solution within 3 h of computing time.

An Offline Framework for Handling Automatic Passenger Counting Raw Data

Knowledge of ridership data on bus routes is pivotal for the quality and efficient operational planning of public transport companies. Automatic passenger counting (APC) can represent a powerful resource for supporting this activity, because it can provide a databank of accurate counts. However, relevant challenges, such as the matching of data to the bus stop, data validation, tackling anomalies, and building intelligible performance reports, must be faced in order to make APC data a mainstream source of information. This paper proposes an offline framework for addressing these challenges. In order to illustrate a possible application of the framework, its use for setting bus frequencies is investigated. The results are represented by easy-to-read control dashboards composed of tables and graphs. The methodology is experimentally tested with data records provided by the bus operator CTM in Cagliari, Italy. Finally, we discuss the implications on service rearrangement.

An optimization model for the short-term manpower planning problem in transhipment container terminals

We study the short-term manpower planning problem for transhipment container terminals.A deterministic optimization model is proposed for this problem.The model solutions are compared to the manpower policy adopted by a real TCT.We show when the terminal policy is effective or there is room for optimization.The model can be optimally solved even in the case of huge transhipment container terminals. This paper investigates the short-term manpower planning problem regarding transhipment container terminals. It consists of determining shifts, tasks and activities of the manpower working in these terminals in order to serve vessels in time intervals, which typically do not overlap with personnel shifts. This complex problem is modelled by an integer linear programming formulation. The optimal solutions of the model are compared with the decisions made in accordance with the manpower policy adopted by a real transhipment container terminal. The experimentation sheds light on when its policy is effective or when there is room for optimisation. The computational tests indicate that the model can be optimally solved even in the case of huge transhipment container terminals.

Optimal management of human resources in transhipment container ports

Human resources allocation plays a key role in transhipment maritime container terminals to achieve high levels of productivity and provide high quality services to shipping companies. The deep interest of container terminals in this issue can be supported by optimization methods. In this work, an optimization model is proposed to determine the optimal daily allocation of crane operators and trailer drivers. Different requirements are taken into account for permanent staff, external workers and personnel shortfall. Since workforce undermanning is a crucial factor for both shipping companies and container terminals, we aim to show that personnel shortfalls and operation delays can be significantly reduced if the model encompasses a longer-than-1-day planning horizon in a rolling horizon fashion.

Heuristics for the routing of trucks with double container loads

This research addresses a problem motivated by a real case study. A carrier must plan the routes of trucks in order to serve importers and exporters. What is original in this vehicle routing problem is the impossibility to separate trucks and containers during customer service and the opportunity to carry up to two containers per truck. Customers may demand more than one container and may be visited more than once. Moreover, according to the carrier’s policy, importers must be served before exporters. In order to address this Vehicle Routing Problem with backhaul and splits, a linear integer programming model is proposed. This research aims to show to what extent an exact algorithm of a state of the art solver can be used to solve this model. Moreover, since some instances are too difficult to solve for the exact algorithm, a number of heuristics is proposed and compared to this algorithm. Finally, the heuristics are compared to the real decisions of the carrier who has motivated this problem.

An Offline Framework for the Diagnosis of Time Reliability by Automatic Vehicle Location Data

Time reliability problems are unavoidable, owing to the stochastic context in which bus services are operated. Therefore, characterizing their reliability and understanding possible sources of unreliability provides an opportunity to keep buses on schedule and/or maintain planned headways. Measuring time reliability is technologically feasible by automatic vehicle location (AVL) systems, which can collect disaggregated data on the delivered service and disclose information on its performance. This paper proposes the first offline framework applicable to any bus route in order to accurately characterize the bus stops and the time periods in which reliability is insufficient, and to disclose the systematic unreliability sources from collected AVL data and select preventive strategies, accordingly. The framework is tested on the real case study of a bus route, using about 40 000 AVL data records provided by the bus operator CTM in Cagliari, Italy. The experimentation shows that this framework can be adopted by transit managers for accurate reliability analysis.

A multi-scenario model for empty container repositioning with uncertain demand

This paper investigates the maritime repositioning of empty containers when their demand is uncertain. It involves moving empty containers from import-dominant ports to export-dominant ports in anticipation of future demands. In order to consider demand uncertainty, the problem is addressed by a stochastic programming approach, in which different scenarios are included in a multi-scenario optimisation model and explicitly linked by non-anticipativity conditions. Numerical experiments are carried out to show which benefits are obtained by multi-scenario solutions with respect to their deterministic counterparts, which consider only one point forecast for each uncertain parameter.