Search

Location decisions are crucial in the spatial organization in both public and private sectors as they can have a long term impact on operational performances and on service levels. Social cost minimization, universality of services and equity, expressed in terms of users' accessibility, are the main objectives in public services contexts. Nevertheless, the enduring trend of public expenditures revision poses, also in the public sectors, the need to pursue objectives of economic efficiency. In the literature, two families of optimization problems are typically used to address these problems, namely Facility Location Problems (FLPs) and Districting Problems (DPs). The aim of this thesis is to show how FLPs and DPs can be used to underpin spatial organization processes of public services, providing analytical models able to assist the decision making. To this end, novel mathematical models are developed with application to the healthcare and postal service sectors. In particular, a hierarchical facility location model is formulated to reorganize an existing regional Blood Management System (BMS) while an integrated location-districting model is proposed for the organization of postal collection operations in urban areas. A constructive heuristic procedure is also devised to solve the latter problem. Extensive computational experiments are realized to validate the proposed models and to show their capability to provide insightful managerial implications. Finally, the thesis aims at filling another existing gap in the literature due to the absence of stochastic models for DPs. Hence, a two-stage stochastic program for districting is introduced and tested on real georgaphic data. Several extensions of the proposed modeling framework are also discussed.

In this paper, a spatial analysis is performed to measure students’ access to the upper secondary education system. Based on the definition of quantitative indicators, the adopted approach is applied to an Italian regional case by exploiting the capabilities of a GIS software and using census tracts’ level data. The obtained results highlight geographical patterns of inequalities in access among students and shed light on the least served areas. Further analysis shows that accessibility reflects the degree of urbanization within the study region and that geographical distances are actual barriers to rural students since they are not compensated for by either economic status or the availability of digital infrastructures. The study offers empirical grounds to inform the decision-making process toward equity-in-access oriented interventions. Longer-term actions, as the activation of new schools (network expansion), the activation of new programs (service expansion) or the redistribution of their supply among the current network (network reorganization), as well as mid-term ones, like offering economic support for students’ mobility, or reinforcing digital connectivity, emerge as relevant to mitigate social exclusion.

Blood is a vital resource in the health care context and then its efficient and effective collection, management and distribution is a fundamental task. As blood is drawn on voluntary basis, an appropriate organization of the collection systems on the territory is crucial; in particular, a widespread presence of collection facilities is important to attract potential donors and assure a given blood collection capacity, but it may have also a strong impact on the total management costs. This study is motivated by a project aimed at reorganizing regional blood management systems in Italy, in order to reduce total management costs without compromising the self-sufficiency goal, i.e. the goal of satisfying the blood demand coming from the region. In particular, we formulate the problem as a facility location model and we apply it using real data related to a specific regional context. The obtained results are analyzed and discussed and then some conclusions are drawn.

In this paper a stochastic districting problem is investigated. Demand is assumed to be represented by a random vector with a given joint probability distribution function. A two-stage mixed-integer stochastic programming model is proposed. The first stage comprises the decision about the initial territory design: the districts are defined and all the territory units assigned to one and exactly one of them. In the second stage, i.e., after demand becomes known, balancing requirements are to be met. This is ensured by means of two recourse actions: outsourcing and reassignment of territory units. The objective function accounts for the total expected cost that includes the cost for the first-stage territory design plus the expected cost incurred at the second stage by outsourcing and reassignment. The (re)assignment costs are associated with the distances between territory units, i.e., the focus is put on the compactness of the solution. The model is then extended in different ways to account for aspects of practical relevance such as a maximum desirable dispersion, reallocation constraints, or similarity of the second-stage solution w.r.t. the first-stage one. The new modeling framework proposed is tested computationally using instances built using real geographical data.

The recent remarkable increase in air passenger traffic has been fostering a considerable congestion of the airport facilities. In this context, traditional procedures employed for check-in operations have been supported by alternative methods, based on the use of self-service options (kiosks, web services, app for mobile phones, etc). However, even if such innovations are contributing to improve the service level provided to passengers, field investigations suggest that traditional procedures will be employed also in the future, especially for medium and long-haul flights, where baggage dropping is required. For this reason, the passengers allocation problem at check-in counters is attracting growing attention by the scientific community and several decision support tools, involving both optimization and simulation methods, have been proposed. Most of the available approaches aim at deciding the optimal number of check-in counters to be activated, in such a way to balance the operative costs and passengers waiting times. Such approaches assume that the service capacity (in terms of available check-in operators and counters) is given and determined on the basis of physical constraints (related to the available space in the terminal) and of staff scheduling decisions made at a tactical level. The present contribution tries to overcome this limitation, by proposing a decision support system, based on a mathematical model, capable of designing optimal check-in policies by also incorporating staff scheduling decisions. The model is tested on some real-world case studies; computational results are evaluated, along with the practical usability of the approach.

Facility Location Models (FLMs) have been widely applied in the context of both private and public sector, to decide the best configuration of new facilities to be located in a given area. In the last years, due to the general interest to reduce costs and improve efficiency, several works focused on problems aimed at modifying the territorial configuration of existing facilities, in terms of number, position and/or capacities, etc. In this work, we propose a new mathematical model to support territorial re-organization decisions in non-competitive contexts. The model assumes the presence of a set of facilities providing different types of services to users (multi-type facilities) and explores the possibility to improve the efficiency of the system by implementing different rationalization actions; i.e., facility closure, service closure, capacity reallocation among services at a given facility. The model aims at finding a trade-off solution between the service efficiency and the need of ensuring a given accessibility level to users. It has been tested on a set of randomly generated instances, to show that a good range of problems can be solved to optimality through the use of a commercial solver (CPLEX).

One of the most important challenges for current supply chains is environmental protection. Companies are beginning to take into account the environmental impact of their operations and make informed decisions on their processes and supply chains, thereby becoming more energy efficient and environmentally sustainable. However, in many cases, especially among SMEs, the related know-how, as well as competences required for successful implementation of the sustainable practices are missing. The aim of this paper is to demonstrate results of a European project that resulted in the development of a process of gaining key competencies for the implementation of energy efficient and sustainable supply chain operations. Furthermore the content discuses and provides an example of an education for sustainable development (ESD) framework, as well as supporting activities in order to enable its successful implementation.

In the general context of welfare reforms in western economies, many actions concerning the rationalization of local administrative structures have been undertaken. In particular, in Italy a recent debate has been addressed about the reduction of the overall number of provinces and the rearrangement of their borders. As provinces are responsible of providing some essential services to the population within their boundaries, any possible scenario should combine the need for more efficient territorial configurations with the safeguard of the services’ accessibility. From a methodological point of view, such problem involves aspects from both facility location and districting problems. In this work, we formulate a mathematical model to support the decision making process and we compare scenarios provided on four benchmark problems, built on the real data associated to the most representative Italian regions.

A cross-dock is a facility where arriving materials are sorted, grouped and delivered to destinations, with very limited storage times, with the overall objective of optimizing the total management costs. The operational efficiency of a cross-docking system strongly depends on how the logistic activities are organized. For this reason, optimization models and methods can be very useful to improve the system performances. In this paper, we propose a mathematical model to describe the so-called truck scheduling problem at a cross-docking platform. The model considers most of the actual constraints occurring in real problems; therefore, it can be viewed as an interesting basis to define a decision support system for this kind of problems. Some preliminary results show that the model can be efficiently solved in limited computational times.