Automotive OEM Demand Forecasting: A Comparative Study of Forecasting Algorithms and Strategies

Jože M. Rožanec,Maja Škrjanc,Dunja Mladenić,Blaž Fortuna,Blaž Kažič

doi:10.3390/app11156787

Jože M. Rožanec, Maja Škrjanc + Show 3 more

Open Access

https://doi.org/10.3390/app11156787

Copy DOI

Abstract

Demand forecasting is a crucial component of demand management, directly impacting manufacturing companies’ planning, revenues, and actors through the supply chain. We evaluate 21 baseline, statistical, and machine learning algorithms to forecast smooth and erratic demand on a real-world use case scenario. The products’ data were obtained from a European original equipment manufacturer targeting the global automotive industry market. Our research shows that global machine learning models achieve superior performance than local models. We show that forecast errors from global models can be constrained by pooling product data based on the past demand magnitude. We also propose a set of metrics and criteria for a comprehensive understanding of demand forecasting models’ performance.

Highlights

Models: on top of Multiple Linear Regression (MLR), support vector regressor (SVR) [56], and multilayer perceptron regressor (MLPR) [57], which we found were used in automotive demand forecasting literature, we evaluate Ridge [58]; Lasso [59]; Elastic Net [60]; K-nearest-neighbor regressor (KNNR) [61]; tree-based regressors (decision tree regressor (DTR), random forest regressor (RFR), and Gradient Boosted Regression Trees (GBRT)); a voting ensemble created using the most promising and diverse algorithms (KNNR, SVR, and RFR); as well as a stacked regression [62] considering
This research compares 21 forecasting techniques to provide future demand estimates for an automotive original equipment manufacturer (OEM) company located in Europe
We considered multiple metrics and criteria to assess forecasting models’ performance

Summary

Introduction

(the “amount of something ready to be used” [1]) and demand The amount that they buy” [1] at a given point in time) are two key elements continually interacting in the market. The ability to accurately forecast future demand enables manufacturers to make operational and strategic decisions on resources (allocation and scheduling of raw material and tooling), workers (scheduling, training, promotions, or hiring), manufactured products (market share increase and production diversification), and logistics for deliveries [2]. Accurate demand forecasts reduce inefficiencies, such as high stocks or stock shortages, which have a direct impact on the supply chain (e.g., reducing the bullwhip effect [3,4]), and prevent a loss of reputation [5]

Objectives

Methods

Results

Conclusion