Inventory policy selection for multi-items and multi-suppliers system using periodic and continuous models

Abstract

Purpose: The purpose of this paper is to select inventory policy model(s) that can minimize total inventory costs considering order lot size, reorder points, lead time, and maximum inventory levels. The evaluation of alternative policies incorporates both service level performance and implementation feasibility considerations. In this paper, the demand is probabilistic and follows normal distribution. Also, the lead time in the study has a potential to be accelerated through controllable replenishment adjustments and decisions. This capability enables the evaluation of trade-offs between replenishment responsiveness and inventory-related costs under probabilistic demand conditions. We consider possible shortages caused by unfulfilled customer demand due to stock shortages to be minimized. The scope of the problem involves multiple suppliers and multi-items.

Design/methodology/approach: Several normal-distributed probabilistic inventory policy models are proposed using perpetual/continuous inventory model (Q) and periodic inventory model (P). Furthermore, we have developed the periodic inventory model (P) for multi-item single-source (MISS) because of the potential for several different items to be supplied by the same supplier. This will be able to minimize the total inventory cost, considering that items are no longer treated individually. In addition, we also investigate the periodic and perpetual/continuous model with varying combination policy of shortage costs (i.e. backorder and lost sales policies).

Findings: The proposed inventory policy is able to save total costs consisting of ordering costs, holding costs, and shortage costs with a value of 68.40% for Supplier 1, 75.14% for Supplier 2, and 41.65% for Supplier 3.  The research results also increased the service level with final service level 99.90% for Supplier 1, 99.91% for Supplier 2, and 97.29% for Supplier 3. The study also showed that the proposed model is sensitive to changes in demand, ordering costs, and holding costs parameters.

Originality/value: This paper offers a new approach, introducing P multi-item model with probabilistic demand in multiple supplier systems considering backorder, lost sales and controllable lead-time acceleration capability. Furthermore, the developed model is compared with existing inventory models to see its effectiveness. The existing models are P-model, Q-model, multi-item model, and lead time reduction model. Of the three suppliers, the integrated P model and multi-item model is the best policy for two suppliers in terms of total inventory cost and service level aspects.