  Citation: Fadda, E.; Perboli, G.; Rosano, M.; Mascolo J.E.; Masera D. A Decision Support System for Supporting Strategic Production Allocation in the Automotive Industry. Sustainability 2022, 14, 2408. https://doi.org/10.3390/su14042408 Academic Editor: Lei Zhang Received: 29 December 2021 Accepted: 16 February 2022 Published: 19 February 2022 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affil- iations. Copyright: © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). sustainability Article A Decision Support System for Supporting Strategic Production Allocation in the Automotive Industry Edoardo Fadda 1, * ,† , Guido Perboli 2,3,† , Mariangela Rosano 2,† , Julien Etienne Mascolo 4 and Davide Masera 4 1 DISMA and ICT for City Logistics and Enterprises Center, Politecnico di Torino, 10129 Turin, Italy 2 DIGEP and ICT for City Logistics and Enterprises Center, Politecnico di Torino, 10129 Turin, Italy; guido.perboli@polito.it (G.P.); mariangela.rosano@polito.it (M.R.) 3 Centre Interuniversitaire de Recherche sur les Reseaux D’entreprise, la Logistique et le Transport , Montreal, QC H3T 1J4, Canada 4 Centro Ricerche Fiat, 10043 Turin, Italy; julien.mascolo@crf.it (J.E.M.); davide.masera@crf.it (D.M.) * Correspondence: edoardo.fadda@polito.it † These authors contributed equally to this work. Abstract: This paper deals with the optimization problem faced by the manufacturing engineering department of an international automotive company, concerning its supply chain design (i.e., deci- sions regarding which plants to open, how many components to produce, and the logistic flow from production to assembly plants). The intrinsic characteristics of the problem, such as stochasticity, the high number of products and components, and exogenous factors, make it complex to formulate and solve the mathematical models. Thus, new decision support systems integrating human choices and fast solution algorithms are needed. In this paper, we present an innovative and successful use case of such an approach, encompassing the decision-maker as an integral part of the optimization process. Moreover, the proposed approach allows the managers to conduct what-if analyses in real-time, taking robust decisions with respect to future scenarios, while shortening the time needed. As a byproduct, the proposed methodology requires neither the definition of a probability distribution nor the investigation of the user’s risk aversion. Keywords: decision support system; stochastic optimization; production allocation; automotive; industrial strategies 1. Introduction Enhancing efficiency, resilience, sustainability, and responsiveness in the supply chain is a relevant issue for automotive companies. Given the complexity of its products, this industry has a complex global supply chain. Indeed, it involves an intricate system of multiple actors engaged in various levels of collaboration (e.g., vehicle manufacturers, components suppliers, third-party contractors). Moreover, some information is not known a priori (e.g., product demand, travel times, duty costs, raw material price, political incen- tives). In this context, one of the biggest challenges faced by this industry is the design of a supply chain that guarantees customer satisfaction, considering the uncertainty of some parameters while containing the production costs. To better face this challenge, companies use the large amount of data they collect with powerful analytical tools [1,2]. Nevertheless, in order to make decisions, the results of statistical analysis are not enough. They must be integrated with the knowledge and insight of the decision-makers (such as tax constraints, marketing decisions, etc.), and properly defined optimization methods. This led to the need to redesign the existing decision support tools through a holistic vision of the system and, thus, a mix of qualitative and quantitative methods. Several effective tools used for decision support rely on mathematical models that represent the problem. However, in a real setting, the resulting problem often becomes so complex that it cannot be solved due to the number of decision-makers involved (drivers, Sustainability 2022, 14, 2408. https://doi.org/10.3390/su14042408 https://www.mdpi.com/journal/sustainability