MODELLAZIONE MATEMATICA DELLE ATTIVITÀ DI MANUTENZIONE E RISANAMENTO STRADALE

The problem addressed in this thesis concerns the planning and scheduling of maintenance activities on the road surface, considering a defined planning horizon. The problem is also addressed by considering the aspect linked to traffic congestion, due to the opening of construction sites in even quite adjacent places. The problem addressed is crucial given that a road network, that operates in poor service conditions, causes an increase in logistics costs. Furthermore, an indiscriminate relocation of road construction sites causes various inconveniences, including an increase in emissions and poor road safety. After a careful examination of the literature, the problem under consideration was modeled mathematically, discretizing the planning horizon into periods, each of equal length, and dividing the road section to be maintained into segments, maximizing a road reliability index. This index takes into account the different maintenance activities selected, in the various periods, for the multiple segments. The innovative aspect, introduced in the proposed formulation, is the modeling of a set of constraints to take into account the fact that it is not possible to select adjacent segments for maintenance in the same period. The model was implemented in Python and solved using the Cplex API. An experimental campaign was therefore conducted as well as a sensitivity analysis as the most significant input parameters of the model varied. The thesis addresses the issue of planning and scheduling maintenance activities on road surfaces within a specific time frame. The focus is on minimizing traffic congestion caused by concurrent construction activities nearby. This problem is significant as poorly maintained road networks lead to increased logistics costs and various inconveniences, including heightened emissions and compromised road safety. The author reviewed existing literature and formulated a mathematical model that divides the planning horizon into equal periods and the road section into segments. The goal is to maximize a road reliability index considering different maintenance activities in various segments within each period. The proposed formulation introduces constraints to prevent the simultaneous selection of adjacent segments for maintenance. The model was implemented in Python and solved using the Cplex API. An experimental campaign and sensitivity analysis were conducted to assess the impact of varying input parameters. The thesis addresses efficient planning and scheduling of road maintenance activities to minimize traffic congestion and associated costs. It introduces a mathematical model to maximize road reliability while preventing simultaneous maintenance of adjacent segments. The model was implemented in Python using the Cplex API and tested through an experimental campaign and sensitivity analysis. The thesis addresses efficient planning and scheduling of road maintenance activities to minimize traffic congestion and associated costs. It introduces a mathematical model to maximize road reliability while preventing simultaneous maintenance of adjacent segments. The model was implemented in Python using the Cplex API and tested through an experimental campaign and sensitivity analysis.

Il problema affrontato in questo lavoro di tesi riguarda la programmazione e la schedulazione delle attività di manutenzione sul manto stradale, considerando un orizzonte di pianificazione definito. Il problema è inoltre affrontato considerando l’aspetto legato alla congestione del traffico, dovuta all’apertura di cantieri in posti anche abbastanza attigui. Il problema affrontato riveste un’importanza cruciale visto che una rete stradale, che opera in scarse condizioni di servizio, provoca un aumento dei costi logistici. Inoltre, un’indiscriminata dislocazione dei cantieri stradali provoca svariati disagi, tra cui un aumento delle emissioni ed una scarsa sicurezza stradale. Dopo un’attenta disamina della letteratura, il problema in esame è stato modellato matematicamente, discretizzando l’orizzonte di pianificazione in periodi, ciascuno di uguale lunghezza e dividendo il tratto stradale da manutenere in segmenti, massimizzando un indice di affidabilità della strada. Questo indice tiene in conto le diverse attività di manutenzione selezionate, nei vari periodi, per i vari segmenti. L’aspetto innovativo, introdotto nella formulazione proposta, è la modellazione di un insieme di vincoli per tenere in conto il fatto che non è possibile selezionare segmenti attigui per la manutenzione nello stesso periodo di tempo. Il modello è stato implementato in Python e risolto tramite le API di Cplex. È stata quindi condotta una campagna sperimentale ed anche un’analisi di sensitività al variare dei parametri di input del modello più significativi.