A Deep Reinforcement Learning Approach for Optimal Replenishment Policy in A Vendor Managed Inventory Setting for Semiconductors Muhammad Tariq Afridi, Santiago Nieto-Isaza, Hans Ehm, Thomas Ponsignon, Abdelgafar Hamed Proceedings Winter Simulation Conference, 2020 Vendor Managed Inventory (VMI) is a mainstream supply chain collaboration model. Measurement approaches defining minimum and maximum inventory levels for avoiding product shortages and over-stocking are rampant. No approach undertakes the responsibility aspect concerning inventory level status, especially in semiconductor industry which is confronted with short product life cycles, long process times, and volatile demand patterns. In this work, a root-cause enabling VMI performance measurement approach to assign responsibilities for poor performance is undertaken. Additionally, a solution methodology based on reinforcement learning is proposed for determining optimal replenishment policy in a VMI setting. Using a simulation model, different demand scenarios are generated based on real data from Infineon Technologies AG and compared on the basis of key performance indicators. Results obtained by the proposed method show improved performance than the current replenishment decisions of the company.
Simulation-optimization approach for the stochastic location-routing problem N Herazo-Padilla, J R Montoya-Torres, S Nieto Isaza, J Alvarado-Valencia Journal of Simulation, 2015 The location routing problem with stochastic transportation cost and vehicle travel speeds is considered in this paper. A hybrid solution procedure based on Ant Colony Optimisation (ACO) and Discrete-Event Simulation (DES) is proposed. After using a sequential heuristic algorithm to solve the location subproblem, the subsequent capacitated vehicle routing problem is solved using ACO. Finally, a DES model evaluates those vehicle routes in terms of their impact on the expected total costs. The approach is tested using well-known randomly generated datasets. Since no previous works in the literature studied exactly the same SLRP, the proposed procedure is compared against its deterministic version. Numerical results show the efficiency and efficacy of the hybrid ACO-DES approach.
A contrast between DEMATEL-ANP and ANP methods for six sigma project selection: A case study in healthcare industry Miguel A Ortíz, Heriberto A Felizzola, Santiago Nieto Isaza BMC Medical Informatics and Decision Making, 2015 BACKGROUND: The project selection process is a crucial step for healthcare organizations at the moment of implementing six sigma programs in both administrative and caring processes. However, six-sigma project selection is often defined as a decision making process with interaction and feedback between criteria; so that it is necessary to explore different methods to help healthcare companies to determine the Six-sigma projects that provide the maximum benefits. This paper describes the application of both ANP (Analytic Network process) and DEMATEL (Decision Making trial and evaluation laboratory)-ANP in a public medical centre to establish the most suitable six sigma project and finally, these methods were compared to evaluate their performance in the decision making process. METHODS: ANP and DEMATEL-ANP were used to evaluate 6 six sigma project alternatives under an evaluation model composed by 3 strategies, 4 criteria and 15 sub-criteria. Judgement matrixes were completed by the six sigma team whose participants worked in different departments of the medical centre. RESULTS: The improving of care opportunity in obstetric outpatients was elected as the most suitable six sigma project with a score of 0,117 as contribution to the organization goals. DEMATEL-ANP performed better at decision making process since it reduced the error probability due to interactions and feedback. CONCLUSIONS: ANP and DEMATEL-ANP effectively supported six sigma project selection processes, helping to create a complete framework that guarantees the prioritization of projects that provide maximum benefits to healthcare organizations. As DEMATEL- ANP performed better, it should be used by practitioners involved in decisions related to the implementation of six sigma programs in healthcare sector accompanied by the adequate identification of the evaluation criteria that support the decision making model. Thus, this comparative study contributes to choosing more effective approaches in this field. Suggestions of further work are also proposed so that these methods can be applied more adequate in six sigma project selection processes in healthcare.
Comparative analysis between ANP and ANP-DEMATEL for six sigma project selection process in a healthcare provider Lecture Notes in Computer Science, 2014
Coupling ant colony optimization and discrete-event simulation to solve a stochastic location-routing problem Nilson Herazo-Padilla, Jairo R. Montoya-Torres, Andres Munoz-Villamizar, Santiago Nieto Isaza, Luis Ramirez Polo Proceedings of the 2013 Winter Simulation Conference Simulation Making Decisions in A Complex World Wsc 2013, 2013 This paper considers the stochastic version of the location-routing problem (SLRP) in which transportation cost and vehicle travel speeds are both stochastic. A hybrid solution procedure based on Ant Colony Optimization (ACO) and Discrete-Event Simulation (DES) is proposed. After using a sequential heuristic algorithm to solve the location subproblem, ACO is employed to solve the corresponding vehicle routing problem. DES is finally used to evaluate such vehicle routes in terms of their impact on the expected total costs of location and transport to customers. The approach is tested using random-generated data sets. because there are no previous works in literature that considers the same stochastic location-routing problem, the procedure is compared against the deterministic version of the problem. Results show that the proposed approach is very efficient and effective.
