Demand forecast modelling of vehicles as a decision support: the case of Toyota Ghana

Abstract

Purpose: This purpose of this paper is to develop a mathematical demand forecast model as an alternative to expert-intensive methods for decision support in automobile companies using Toyota Ghana as a case. The paper explores the challenges associated with reliance on experts’ judgment in demand forecasting.
Design/Methodology/Approach: The methodology involved analysing stock reports, lost sales reports, and financial reports from Toyota Ghana to understand the effect of poor forecasting. Using data from two key managers and six sales staff, the project examines the perspectives of staff regarding the use of expert judgment for demand forecasting. Further data was collected via a questionnaire from five authorized automobile distributors and dealerships.
Findings: The results revealed the adverse effects of expert-opinion forecasting, which include irregular stock quantities leading to lost sales, vehicle quality challenges leading to deterioration, and long-term negative impact on profitability. Yet demand forecasting by reliance on experts was very prevalent in the automobile industry. The developed forecast model relies on Mean Absolute Percentage Error with a smoothing constant of 0.4. was validated using recent historical data revealing a 2% variance with actual demand values, while for expert judgment the variation margin was 14%. This strongly indicated that the model yielded more accurate predictions of demand than expert predictions.
Research Limitation: The case-study nature of the study means a more generalized study was still needed before the findings could be more widely applied across the automobile industry.
Practical implication: The study recommended further development of scientific forecasting models for predicting demand across the automobile industry since they carried positive implications for the smooth running of the industry. This could help mitigate the challenges associated with using expert opinions in demand forecasting. Beyond this, the model could serve to provide valuable information to vehicle manufacturers, thereby yielding efficiencies in their value chains.
Social implication: Accurate demand forecasting and management have positive implications for operational efficiency that minimizes customer disappointment.
Originality / Value: The model offers a better alternative for predicting demand more accurately, promoting correct stock holding quantities, avoiding stock deterioration, and reducing expenditure on quality checks, thus ultimately increasing profitability.