Validation of Inventory models for Single-echelon Supply Chain using Discrete-event Simulation

TL;DR

This paper evaluates the accuracy of inventory models for single-echelon supply chains using discrete-event simulation, revealing conditions where models under-predict service levels and others where they align well.

Contribution

It assesses the applicability of existing inventory frameworks through simulation, highlighting their limitations and accuracy in various scenarios.

Findings

Models often under-predict service levels

Some scenarios show statistically significant model accuracy

Simulation reveals discrepancies in safety stock predictions

Abstract

Inventory decision frameworks proposed in the literature for single-echelon supply chain systems rely on assumptions to obtain closed form expressions. In particular, two such frameworks - one conventional and the other with a demand undershoot - determine optimal reorder point for a desired $\beta$ or Type-II service level. In this work we assess the accuracy and applicability of these frameworks with the help of a discrete event simulation framework developed in SimPy. While in several cases the closed form literature models under-predict the service level, and thus result in a higher reorder point and safety stock, we observe some situations where model predicted service levels match the simulated results with statistical significance.