A hybrid model based on dynamic programming, neural networks, and surrogate value for inventory optimisation applications

Abstract

This paper proposes a new approach to minimise inventory levels and their associated costs within large geographically dispersed organisations. For such organisations, attaining a high degree of agility is becoming increasingly important. Linear regression-based tools have traditionally been employed to assist human experts in inventory optimisation; endeavours; recently, Neural Network (NN) techniques have been proposed for this domain. The objective of this paper is to create a hybrid framework that can be utilised for analysis, modelling and forecasting purposes. This framework combines two existing approaches and introduces a new associated cost parameter that serves as a surrogate for customer satisfaction. The use of this hybrid framework is described using a running example related to a large geographically dispersed organisation.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 20 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.