Title: A smart DDMRP model using machine learning techniques

Authors: Jose Aguilar; Ricardo José Dos Santos Guillén; Rodrigo García; Carlos Gómez; M. Jerez; Marvin Luis Jiménez Narváez; Eduard Puerto

Addresses: CEMISID Faculta de Ingeniería, Universidad de Los Andes, Mérida, Venezuela; GIDITIC, Universidad EAFIT, Medellín, Colombia ' CEMISID Faculta de Ingeniería, Universidad de Los Andes, Mérida, Venezuela ' GIDITIC, Universidad EAFIT, Medellín, Colombia; Facultad de Ciencias e Ingenierías, Universidad de Sinú, Montería, Colombia ' EXEK Company, Medellín, Colombia ' CEMISID Faculta de Ingeniería, Universidad de Los Andes, Mérida, Venezuela ' Facultad de Ciencias e Ingenierías, Universidad de Sinú, Montería, Colombia ' Grupo GIA, Universidad Francisco de Paula Santander, Cúcuta, Colombia

Abstract: This paper proposes a hybrid algorithm based on the demand-driven manufacturing resources planning (DDMRP) model and machine learning techniques to determine when and how much to purchase a product. The DDMRP model optimises the inventory using predictive models to determine the product demands, and the behaviour of the providers. Then, our DDMRP model determines when and how much to purchase. Thus, our approach defines a smart inventory management to establish what should be purchased and when. The preliminary results are very encouraging because the inventory follows the optimal levels by product based on demand, avoiding a lack of inventory.

Keywords: inventory management; demand-driven model; machine learning; supply chain; DDMRP.

DOI: 10.1504/IJVCM.2023.130973

International Journal of Value Chain Management, 2023 Vol.14 No.2, pp.107 - 142

Received: 12 Oct 2021
Accepted: 30 Jan 2022
Published online: 17 May 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article