Politecnico di Torino (logo)

Industrial Vehicles’ Inventory Management using Data-Driven Stochastic Optimization and Forecasting

Giorgia Guarnotta

Industrial Vehicles’ Inventory Management using Data-Driven Stochastic Optimization and Forecasting.

Rel. Luca Vassio, Marco Mellia. Politecnico di Torino, Corso di laurea magistrale in Ict For Smart Societies (Ict Per La Società Del Futuro), 2022

PDF (Tesi_di_laurea) - Tesi
Licenza: Creative Commons Attribution Non-commercial No Derivatives.

Download (3MB) | Preview

Warehouse optimization is a process that aims at improving the management of time, space, and resources inside a warehouse, minimizing the overall costs yet ensuring a satisfactory quality of service for the customers. When running an efficient warehouse most effort is on the inventory management, which is a well-known challenge for businesses: on the one hand they have to avoid late product supply, which would result in lost profits, while on the other hand they need to focus on a careful inventory control to mitigate the rise of costs, caused by an excessive accumulation of products. The purpose of this thesis is that of supporting industrial fleet managers when vehicles undergo maintenance, providing a way to cope with the need of spare parts under realistic situations, represented by the uncertain nature of components demand. This work proposes a Two-stage Stochastic Mixed-Integer Nonlinear Problem that aims at minimising both spare parts inventory costs and vehicle offline periods when the requested items are not immediately available. The presence of historical data allows to implement a data-driven approach: we use data collected between 2020 and 2022 from HINO Motors maintenance history to investigate the spare parts demand distributions and provide an efficient forecast. First, we investigate the case in which the demand distribution is stationary and does not change over time, providing an optimal target inventory level to maintain for ensuring an efficient warehouse management. Secondly, we study the possibility that the demand may vary over different time periods, thus providing the optimal order quantity per spare part and keeping into account the difference in the shipment time per component. Finally, we also exploit machine learning techniques, introducing the uncertainty of forecast of spare parts demand in the optimization process. Results show that, through automation and a careful stock control strategy, businesses are able to improve customer satisfaction and reduce their inventory holding costs, especially if compared with manual warehouse management or naive strategies.

Relators: Luca Vassio, Marco Mellia
Academic year: 2022/23
Publication type: Electronic
Number of Pages: 113
Corso di laurea: Corso di laurea magistrale in Ict For Smart Societies (Ict Per La Società Del Futuro)
Classe di laurea: New organization > Master science > LM-27 - TELECOMMUNICATIONS ENGINEERING
Aziende collaboratrici: UNSPECIFIED
URI: http://webthesis.biblio.polito.it/id/eprint/25466
Modify record (reserved for operators) Modify record (reserved for operators)