Title: Managing Electric Vehicle and Drone Battery Swap Stations Under Uncertainty - given by Amin Asadi
- Abstract: Electric vehicles (EVs) and drones hold great promise for revolutionizing transportation and supply chains. However, battery-oriented issues, including range anxiety and battery degradation, impede adoption. Battery swap stations (BSSs) are one alternative to reduce these concerns by allowing quick swapping of depleted for full batteries. This research presents the stochastic scheduling, allocation, and inventory replenishment problem (SAIRP) for a BSS when explicitly considering the uncertain arrival of swap demand, highly-variable charging costs, battery degradation, and replacement. We model the stochastic SAIRP using a finite horizon Markov Decision Process (MDP) to determine when and how many batteries are charged, discharged, and replaced over time. We show that SAIPRs suffer from the curses of dimensionality. Therefore, we provide intelligent Approximate Dynamic Programming (ADP) and Reinforcement Learning (RL) methods that benefit from the mathematical structure of the problem. We prove the monotonicity of the value function in general and the monotonicity of an optimal policy in special cases of SAIRPs. In computational tests, we demonstrate the superior performance of our proposed ADP/RL methods as compared to exact methods and other approximate solution methods. Further, with the tests, we deduce insights for managing operations in a BSS.
- Bio: Amin Asadi is an Assistant Professor at the Industrial Engineering and Business Information Systems Department at the University of Twente (UT). He received his Ph.D. in Industrial Engineering from the University of Arkansas (U of A). He is the recipient of five awards from the Arkansas Academy of Industrial Engineering (AAIE) from 2018 to 2020 and the Cum laude Award from INFORMS as the president of U of A INFORMS student chapter in 2021. Amin's research interests are in the theory and applications of operations research, artificial intelligence, reinforcement learning (RL), and machine learning for analyzing complex systems and providing optimal/near-optimal decisions for problems dealing with uncertainty. Since joining UT, he started to collaborate with the RL and CHOIR research groups to apply his knowledge and advance in his favorite research domains: transportation, logistics, and Healthcare.
Title: The role of Composite Structural Equation Modelling and ArchiMate in improving social implications of an IT artefact - given by Iqbal Mukti
- Abstract: This presentation will discuss the design process of an IT artefact based on the design science research methodology that considers the social aspects that are likely to be impacted. The design process emphasises the use of the composite structural equation modelling (cSEM) to understand the social artefact, and ArchiMate to translate the understanding of the social artefact into the design of an IT artefact. This design process has been applied in designing a rural smartness platform, a digital platform tailored to reduce the economic inequalities between rural and urban areas, particularly in the developing countries.
- Bio: Iqbal Yulizar Mukti is currently a Ph.D. candidate within the Department of Industrial Engineering and Business Information System, University of Twente, the Netherlands, with a scholarship awarded from the Indonesia Endowment Fund for Education (LPDP). He holds a master degree in Industrial Engineering & Management and a bachelor degree in Mathematics from Institute Teknologi Bandung (ITB), Indonesia. His current research focuses on the adoption of smartness in the rural context, in particular with the goal to improve the rural economic welfare.