Reinforcement Learning Approach for Efficient Inventory Policy in Multi-Echelon Supply Chain Under Various Assumptions and Constraints

Background: Inventory policy highly influences Supply Chain Management (SCM) process. Evidence suggests that almost half of SCM costs are set off by stock-related expenses.Objective: This paper aims to minimise total inventory cost in SCM by applying a multi-agent-based machine learning called Reinforcement Learning (RL).Methods: The ability of RL in finding a hidden pattern of inventory policy is run under various constraints which have not been addressed together or simultaneously in previous research. These include capacitated manufacturer and warehouse, limitation of order to suppliers, stochastic demand, lead time uncertainty and multi-sourcing supply. RL was run through Q-Learning with four experiments and 1,000 iterations to examine its result consistency. Then, RL was contrasted to the previous mathematical method to check its efficiency in reducing inventory costs.Results: After 1,000 trial-error simulations, the most striking finding is that RL can perform more efficiently than the mathematical approach by placing optimum order quantities at the right time. In addition, this result was achieved under complex constraints and assumptions which have not been simultaneously simulated in previous studies.Conclusion: Results confirm that the RL approach will be invaluable when implemented to comparable supply network environments expressed in this project. Since RL still leads to higher shortages in this research, combining RL with other machine learning algorithms is suggested to have more robust end-to-end SCM analysis. Keywords: Inventory Policy, Multi-Echelon, Reinforcement Learning, Supply Chain Management, Q-Learning

Exploring risk pooling in hospitals to reduce demand and lead time uncertainty

Nearly every eighth German hospital faces an elevated risk of bankruptcy. An inappropriate use of inventory management practices is among the causes. Hospitals suffer from demand and lead time uncertainty, and the current COVID-19 pandemic worsened the plight. The popular business logistics concept of risk pooling has been shown to reduce these uncertainties in industry and trade, but has been neglected as a variability reduction method in healthcare operations research and practice. Based on a survey with 223 German hospitals, this study explores how ten risk pooling methods can be adapted and applied in the healthcare context to reduce economic losses while maintaining a given service level. The results suggest that in general risk pooling may improve the economic situation of hospitals and, in particular, inventory pooling, transshipments, and product substitution for medications and consumer goods are the most effective methods in the healthcare context, while form postponement may be unsuitable for hospitals due to the required efforts, delay in treatments, and liability issues. The application of risk pooling in healthcare requires willingness to exchange information and to cooperate, adequate IT infrastructure, compatibility, adherence to healthcare laws and regulations, and securing the immediate treatment of emergencies. Compared to manufacturing and trading companies, hospitals seem to currently neglect the variability reducing effect of risk pooling.

Production -Marketing Coordination under Demand and Lead -Time Uncertainty

In this paper, we consider a make-to-order supply chain which satisfies demand that is dependent on both price and quoted lead -time. The manufacturer chooses the lead -time and the order quantity, and the retailer sets the revenue shares. The interactions between the manufacturer and the retailer are modelled as a Nash Game, and the existence and uniqueness of pure strategy equilibrium are demonstrated. A mechanism that enables the supply chain to coordinate the decisions of the members is developed. Lastly, we also analyze how the supply chain system parameters impact the optimal supply chain decisions and the supply chain performance.

ANALISA PERBAIKAN KINERJA PENGIRIMAN PRODUK R754046 DI PT PELANGI ELASINDO DENGAN PENDEKATAN SAFETY STOCK

PT Pelangi Elasindo is one of the leading manufacturers in Indonesia specializing in webbing production, shoelaces, cords, high quality Molded Pulp, elastics and tape. As one of the vendors for leading shoe factories in the world, the accuracy of shipping goods both in terms of quantity and time is very important because this is a benchmark for calculating shipping performance. There was a delay in delivery at PT Pelangi Elasindo on the R7540460000 Molded Cardboard Shape size 2 product, making delivery performance decrease. Delivery performance in January is 93% while the minimum delivery performance standard is 95%. This study aims to analyze the factors that delay the delivery of goods to customers and improve delivery performance. Based on the results of the analysis, the factors that cause the delay in delivery are damage to the machine, manual cutting and demand variations and uncertainty of lead time. Among these factors, demand variations and lead time uncertainty are the most influential factors, so to improve delivery performance, safety stock must be prepared to avoid stock outs and delivery delays. Based on the results of the calculation of the safety stock that must be prepared is 8,403 prs per week. Key words : Delivery, performance, safety stock,improve

Supplier Replacement Model in a One-Level Assembly System under Lead-Time Uncertainty

Supplier selection/replacement strategies, purchasing price negotiation and optimized replenishment policies play a key role in efficient supply chain management in today’s dynamic market. Their importance increases even more in Industry 4.0. In this paper, we propose a joint model of replenishment planning and purchasing price negotiation in the context of supplier replacement in a one-level assembly system (OLAS) producing one type of finished product. The real component lead times are stochastic. There is consequently a non-negligible risk that the assembly process may be stopped if all components for assembly are not delivered on the due date. This incurs inventory-related costs, holding and backlogging, which should be minimized. We consider a set of suppliers characterized by their prices and the probability distributions of their lead-times, and we present a model and an approach that optimize not only replenishment policy, but also purchasing prices. For a given unit, it is possible to model several alternative suppliers with alternative pricing and lead-time uncertainties, and evaluate their impacts on the total cost: composed of holding, backlogging and purchasing costs for the assembly system. The findings of this study indicate that it can be beneficial to pay suppliers an additional purchase cost in order to reduce the holding and backlogging costs related to uncertainty. In consequence, decision makers can use the proposed approach to negotiate prices and delivery delays or to select suppliers.

Supplier Replacement Model in a One-Level Assembly System under Lead-Time Uncertainty

Supplier selection/replacement strategies and optimized purchasing policies play a key role in efficient supply chain management in today’s dynamic market. Here we study supplier replacement in a one-level assembly system (OLAS) producing one type of finished product. To assemble the product, we need to provide multi-type components, but assembly will be interrupted if any single component is missing, and incoming units will get hoarded until the missing component arrives. The assembly process can be interrupted by various sources of uncertainty, including delays in component deliveries. There is consequently a non-negligible risk that the assembly process may get stopped any moment. This brings inventory-related costs, which should be minimized. Here we consider discrete lead-time distributions to mimic industry-world reality. We present a model that takes into account not only optimal assignment of component order release dates but also replacement of a critical supplier. For a given unit, we model several alternative suppliers with alternative pricing and lead-time uncertainties, and we evaluate the impact on the total assembly system. For a more general case where several suppliers may be replaced, we propose a genetic algorithm.