Advanced OR and AI Methods in Transportation LOGISTICS TERMINALS PLANNING FOR AN OPTIMAL FREIGHT MOBILITY SYSTEM: AN APPLICATION ON A REGIONAL SCALE

This paper suggests an approach to plan logistics terminals. In particular, it describes an application to the freight transport by road involving the sicilian area. The modelling framework, which is inspired by the “Stackelberg game” paradigm, uses an objective function representing some relevant public interests and simulates the choice behaviour of freight transport operators. Basic assumptions concern the involvement of the public sector in terms of a share in investments and incentives to foster the use of terminals; for some scenarios on the public budget constraints, a set of optimal location patterns is determined. 1. Problem definition Inside Sicily, the short distances separating freight generations from attractions have promoted the development of road haulage: in 2001, the 94% of the inland freight traffic uses the road mode. As for the Sicily-External Area non-oil traffic, in 2001, the road travel demand shows the significant incidence of 30 %. The main features of the road freight transport in Sicily have been highlighted combining the data base resulting from a recent survey concerning Italy (source: Italian Trucker Register, 2000) and the outcome of meetings with managers of the main sicilian associations of freight movers: great presence of small-sized firms, which implies a weak propensity to invest in advanced technologies and low load factors; lack of areas for parking vehicles under good conditions of safety; there is no sanitary control and an inadequate packaging for local products coming of the primary sector, which play a keyrole for sicilian exports towards north italian and european markets. In order to make the road-based transport more efficient and the regional productive system more competitive, sicilian policy makers have examined the possibility of creating 1 Department of Aeronautic and Transportation Engeneering at the University of Palermo, Viale delle Scienze Parco d’Orléans 90128 Palermo, [email protected] 2 Department of Aeronautic and Transportation Engeneering at the University of Palermo, Viale delle Scienze Parco d’Orléans 90128 Palermo, [email protected] Logistics terminals planning for an optimal freight mobility system... 161 logistics facilities. Thus, also by implementing Intelligent Transport Systems, the transportation process could be optimized, as regards load factors (consolidation of loads bound for the same destination), path costs, environmental impacts, and several services for goods could be offered like storage, processing, assembling, packaging, quality checks, ... So, the research presented here faces the following challenging questions: how many terminals should be built in Sicily? Where? How much of the considered freight traffic could this logistics network capture? Is it advisable for public administrators (national, regional level) to provide incentives for transhipment? 2. Survey of literature and model framework The present work is based on different research trends. Firstly, readers have to learn that, generally speaking, facility location problems have been tackled by applying the methodologies of operations research. Crainic and Laporte (see [2]) classify the main location models as follows: covering models, that place facilities at the vertices of a network so as to minimize facility costs and to make all the remaining vertices be within a maximal distance from a facility; center models, that locate facilities at the vertices of a network, minimizing the maximal vertex-facility distance; median models, that locate facilities at the vertices of a network and allocate demand, minimizing the total weighted distance between facilities and demand points. Secondly, the present work is inspired by von Stackelberg’s studies about the leaderfollower approach (see [5]). The economist von Stackelberg introduced a model in which one of the competing enterprises in a duopolistic market (leader) carries out its own choice first, taking into account the reaction of the other economic agent (follower) so as to provide this one with the input for its decisions. Thirdly, it is opportune to mention the research activity about discrete choice behaviour modelling through the random utility theory. This postulates that an alternative i will be preferred to an option j if i maximizes the decision maker utility function, whose formulation is probabilistic, since it cannot be determined with accuracy by the modeller (see [1], [3], [4]). The approach illustrated in this paper can be defined as a two player extensive game with perfect information. In detail, it consists of two levels of problem: • the strategic or leader level, relative to the planner choice dimension. The object is maximizing the welfare function including the following elements: facility users’ surplus, facility costs (terminal construction) chargeable to the public sector, public expenditure to favour transhipment. Decision variables: location pattern of terminals and public incentives. • The tactical or follower level, relating to the facility users (transport operators) choice behaviour. The aim is maximizing the shipment utility function, depending on the following elements: trasportation monetary cost, travel time, waste of time at terminal (function of demand to simulate congestion), price for using the considered set of logistics services. Decision variables: choice between the “No transhipment” option and the “transhipment” one, terminal choice, route choice. In particular, the model riproduces the different decision processes according to a “topdown” hierarchy: the planner sets the optimal pattern in terms of terminal locations and 162 M. Catalano, M. Migliore incentive policy, taking account of the transport operators reactions and the resulting goods movements on the transport network. 3. Mathematical formulation 3.1. The leader problem