Forecasting Industrial Wastes to Wastes Disposal Management by Using Box- Jenkins Autoregressive Integrated Moving Average Models and Excel Application: Case Study on Opthalmic Plastic Lens Production of Company A in Thailand

摘要

The purpose of this study is to develop forecasting models for four kinds of wastes (Contaminated materials, Monomers, Used solvents and Wastewater) and apply the outputs of forecasts to an Excel application to plan, implement and control the assets, physical facilities and money investments to support the wastes disposal and transportation of both Company A and four service providers. The method selected uses Box-Jenkins method with data periods from January 2008 to December 2016 (108 series data). After studying these data (Four waste types) using Minitab, fitted models for generating best forecasting values are ARIMA (1, 0, 1) for Contaminated Materials waste, ARIMA (1, 0, 0) or AR (1) for Monomer waste, ARIMA (1, 0, 2) or ARMA (1, 2) for Used Solvents waste and ARIMA (1, 1, 0) or ARI (1, 1) for Wastewater. The results of forecasting the wastes in Company A had RMSE (Root Mean Square Error) (0.388, 0.047, 0.060 and 0.043 respectively) lower than another research paper (1.305). For suitable forecasting models, these models can generate valuable forecasts for the company and its service providers to utilize their budget of money, assets and facilities in Excel application.