An Empirical Comparison of the Sales Forecasting Performance for Plastic Tray Manufacturing Using Missing Data

Che-Yu Hung,Bernard C Jiang,Shi-Woei Lin,Chien-Chih Wang

doi:10.3390/su14042382

Che-Yu Hung, Bernard C Jiang + Show 2 more

Open Access

https://doi.org/10.3390/su14042382

Copy DOI

Abstract

The problem of missing data is frequently met in time series analysis. If not appropriately addressed, it usually leads to failed modeling and distorted forecasting. To deal with high market uncertainty, companies need a reliable and sustainable forecasting mechanism. In this article, two propositions are presented: (1) a dedicated time series forecasting scheme, which is both accurate and sustainable, and (2) a practical observation of the data background to deal with the problem of missing data and to effectively formulate correction strategies after predictions. In the empirical study, actual tray sales data and a comparison of different models that combine missing data processing methods and forecasters are employed. The results show that a specific product needs to be represented by a dedicated model. For example, regardless of whether the last fiscal year was a growth or recession year, the results suggest that the missing data for products with a high market share should be handled by the zero-filling method, whereas the mean imputation method should be for the average market share products. Finally, the gap between forecast and actual demand is bridged by employing a validation set, and it is further used for formulating correction strategies regarding production volumes.

Highlights

To deal with increasingly fierce market competition, manufacturers have transformed their policies by providing customers with customized products and services, quickly responding to diversified needs, reducing competition uncertainty, and obtaining satisfactory services
A careful observation of Table 6 reveals that regardless of the forecaster it is combined with, the zero-filling method almost achieved the best performance in the model evaluation and deployment
It reveals that that regardless regardlessof ofthe theforecaster forecasteritit was was combined combined with, with, the the zero-filling zero-filling method method almost almost achieved achieved the best performance in the model evaluation and deployment

Summary

Introduction

To deal with increasingly fierce market competition, manufacturers have transformed their policies by providing customers with customized products and services, quickly responding to diversified needs, reducing competition uncertainty, and obtaining satisfactory services. Manufacturers expect to maintain or even increase their sales through such a transformation under a potentially increasing inventory pressure. As the green production and the circular economy have gradually formed a consensus between production and sales, manufacturers have tried to address the above challenges and turn them into a positive force to solve market uncertainty and effectively manage their inventory of existing production models. The use of time series models to assist business decision-making has proven its success in many sectors and industries, such as energy consumption forecasting in the petrochemical industry [13,14], station expansion and capacity growth forecasting in the bus system [15], and economic and financial growth forecasting [16].

Objectives

Methods

Results

Discussion

Conclusion