Inventory Management in Multi-Product, Multi-Demand Disassembly Line using Reinforcement Learning
نویسندگان
چکیده
Disassembly lines are the best way to disassemble products with similar components in large quantities. A disassembly line consists of a series of workstations. However, unlike their assembly counterpart, disassembly lines are affected by factors such as multiple component demand arrivals and end-of-life (EOL) product arrivals, uncertainty in EOL product and component demand arrivals and varying processing times. Altogether, these factors make a disassembly line very complex and cause demand fluctuations in the inventory levels of sub-assemblies and components in the system. The fluctuations in the inventory levels lead to increased cost of operation and decreased customer satisfaction. This research focuses on addressing inventory management problem in multi-product, multidemand disassembly line using Reinforcement Learning, specifically Q-learning, which learns from experience to perform very effectively in dynamic environments with stochastic elements.
منابع مشابه
Presenting a Multi Objective Model for Supplier Selection in Order to Reduce Green House Gas Emission under Uncertion Demand
Recently, much attention has been given to Stochastic demand due to uncertainty in the real -world. In the literature, decision-making models and suppliers' selection do not often consider inventory management as part of shopping problems. On the other hand, the environmental sustainability of a supply chain depends on the shopping strategy of the supply chain members. The supplier selection pl...
متن کاملCase-based reinforcement learning for dynamic inventory control in a multi-agent supply-chain system
Reinforcement learning (RL) appeals to many researchers in recent years because of its generality. It is an approach to machine intelligence that learns to achieve the given goal by trial-and-error iterations with its environment. This paper proposes a case-based reinforcement learning algorithm (CRL) for dynamic inventory control in a multi-agent supply-chain system. Traditional time-triggered...
متن کاملA Stochastic Programming Approach for a Multi-Site Supply Chain Planning in Textile and Apparel Industry under Demand Uncertainty
In this study, a new stochastic model is proposed to deal with a multi-product, multi-period, multi-stage, multi-site production and transportation supply chain planning problem under demand uncertainty. A two-stage stochastic linear programming approach is used to maximize the expected profit. Decisions such as the production amount, the inventory level of finished and semi-finished product, t...
متن کاملA Multi-Periodic Multi-Product Inventory Control Problem with Discount: GA Optimization Algorithm
In this article, a finite horizon, multi product and multi period economic order quantity like seasonal items is considered where demand rate is deterministic and known but variable in each period. The order quantities of items come in batch sizes and the end of the period order quantity and, consequently, demand of customers are zero. In addition, storage space is constrained and the problem...
متن کاملA Multi Objective Geometric Programming Model for Optimal Production and Marketing Planning
This paper presents a multi objective geometric programming model which determines the product`s selling price in two markets. We assume demand is a function of price and marketing expenditure in two markets. The cost of production is also assumed to be a function of demands in both markets. Our model is a posynomial function which is solved using Geometric Programming (GP). In our GP implement...
متن کامل