University of Twente Student Theses

Login

Improving Inventory Management at Versuni : Machine Learning for Inventory Forecasting and Dynamic Lot Size Model

Lestario, Gregorius Ugohari (2024) Improving Inventory Management at Versuni : Machine Learning for Inventory Forecasting and Dynamic Lot Size Model.

[img] PDF
1MB
Abstract:This thesis, conducted as part of a Bachelor’s degree in Industrial Engineering and Management during an internship at Versuni, addresses significant issues with inventory forecasting and order quantities at the company. Versuni, a leader in domestic appliances, struggles with inventory forecasts that often deviate from actual levels due to inaccurate demand predictions influenced by dynamic trends, economic changes, and rapid growth. The research focused on improving forecast accuracy and avoiding stockouts by developing machine learning models and optimizing order quantities. The analysis revealed that Versuni's static forecasting formula and automated ERP ordering process led to frequent stockouts and ineffective order quantities. To address this, machine learning models, specifically Linear Regression and Neural Networks, were developed. The Neural Network outperformed others with a 76% reduction in mean squared error (MSE) and a 26% reduction in mean absolute error (MAE), increasing the R-squared value from 76% to 93%. For order quantity optimization, a deterministic dynamic lot size model was implemented in VBA, effectively reducing total inventory costs for 7 of 9 products and preventing stockouts. If stockout costs were included, the model would reduce costs for 8 of 9 products.
Item Type:Essay (Bachelor)
Clients:
Versuni, Amsterdam, Netherlands
Faculty:BMS: Behavioural, Management and Social Sciences
Subject:01 general works
Programme:Industrial Engineering and Management BSc (56994)
Link to this item:https://purl.utwente.nl/essays/103079
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page