Intelligent Waste Management in Industry 4.0

Dai Kaiwen

Abstract

Solid waste output in Singapore has been increased significantly over the years, this results in an increasing need for a well-planned waste collection process. With the implementation of IoT device for waste tracking in industry 4.0, real-time waste relevant information can be gathered and further processed to assist in waste collection management. This project aims at providing an approach on intelligent waste management in industry 4.0 for the waste collection center to make better waste collection decisions.

Predictive models are built based on data collected from the IoT device to make forecasts on future waste generation. Exponential Smoothing method, Auto Regressive Integrated Moving Average method and Artificial Neural Networks method are used to build the forecasting models. The predicted waste amount is then analyzed to determine the optimal waste collection schedule for each collection point. Based on the collection schedule, the waste collection route can be optimized by solving the formulation of an operation research model. The optimized collection schedule and route will then be updated automatically on a daily basis. In this way, an intelligent waste management process in industry 4.0 can be implemented to provide the waste collection center with better waste collection decisions.

There are some limitations in the forecasting models and the operation research models. For the forecasting models, the accuracy level is limited due to limited amount of data available for model building and testing, more data is needed to provide more reliable forecasts. To make more accurate forecasts, dummy variables can be added to model other external effects. As for the waste collection management model, the scope of application for the model to work well is limited due the assumptions made, more details should be considered in the model to satisfy the specific situation for different cases. To implement better intelligent waste management process in industry 4.0, continuous improvements are needed in the future.