Dottorato in INGEGNERIA INDUSTRIALE E GESTIONALE

Titolo della tesi: Methodologies for the Assessment of Industrial and Energy Assets Based on Data Analysis and BI

In July 2020, post pandemic onset, Europe launched the Next Generation EU (NGEU) program. The amount of resources deployed to revitalize Europe has reached 750 billion. The NGEU initiative directs significant resources to Italy. These funds can enable our country to boost investment and increase employment. The missions of Italian Recovery and Resilience Plan (PNRR) include digitization, innovation and sustainable mobility (rail network investments, etc.). In this context, this doctorate thesis discusses the importance of infrastructure for society with a special focus on energy, railway and motorway infrastructure. The central theme of sustainability, defined by the World Commission on Environment and Development (WCDE) as ''development that meets the needs of the present generation without compromising the ability of future generations to meet their needs’’, is also highlighted. Through their activities and relationships, organizations contribute positively or negatively to the goal of sustainable development. Sustainability becomes an integrated part of corporate culture. First research in this thesis describes how Artificial Intelligence techniques can play a supporting role for both maintenance operators in tunnel monitoring and those responsible for safety in operation. Relevant information can be extracted from large volumes of data from sensor equipment in an efficient, fast, dynamic and adaptive manner and made immediately usable by those operating machinery and services to support rapid decisions. Performing sensor-based analysis in motorway tunnels represents a major technological breakthrough that would simplify tunnel management activities and thus the detection of possible deterioration, while keeping risk within tolerance limits. The idea involves the creation of an algorithm for detecting faults, acquiring real-time data from tunnel subsystem sensors and using it to help identify the tunnel's state of service. Artificial intelligence models were trained over a six-month period with a granularity of one-hour time series measured on a road tunnel forming part of the Italian motorway systems. The verification was carried out with reference to a series of failures recorded by the sensors. The second research argument is relates to the transfer capacities of high-voltage overhead lines (HV OHL), which are often limited by the critical temperature of the power line, which depends on the magnitude of the current transferred and the environmental conditions, i.e. ambient temperature, wind, etc. In order to use existing power lines more effectively (with a view to progressive decarbonization) and more safely with respect to critical power line temperatures, this work proposes a Dynamic Thermal Rating (DTR) approach using IoT sensors installed on a number of HV OHL located in different geographical locations in Italy. The objective is to estimate the temperature and ampacity of the OHL conductor, using a data-driven thermo-mechanical model with a bayesian probabilistic approach, in order to improve the confidence interval of the results. This work shows that it might be possible to estimate a spatio-temporal temperature distribution for each OHL and an increase in the threshold values of the effective current to optimize the OHL ampacity. The proposed model was validated using the Monte Carlo method. Finally, in this thesis is presented study on KPIs as indispensable allies of top management in the asset control phase. They are often overwhelmed by the availability of a huge amount of Key Performance Indicators (KPIs). Most managers struggle In understanding and identifying the few vital management metrics and instead collect and report a vast amount of everything that is easy to measure. As a result, they end up drowning in data, thirsty for information. This condition does not allow good systems management. The aim of this research is help the Asset Management System (AMS) of a railway infrastructure manager using business intelligence (BI) to equip itself with a KPI management system in line with the AM presented by the normative ISO 55000 - 55001 - 55002 and UIC (International Union of Railways) guideline, for the specific case of a railway infrastructure. This work starts from the study of these regulations, continues with the exploration, definition and use of KPIs. Subsequently KPIs of a generic infrastructure are identified and analyzed , especially for the specific case of a railway infrastructure manager. These KPIs are fitted in the internal elements of the AM frameworks (ISO-UIC) for systematization. Moreover, an analysis of the KPIs now used in the company is made, compared with the KPIs that an infrastructure manager should have. Starting from here a gap analysis is done for the optimization of AMS.