Prediction of MSW in Delhi: Application of sARIMA model

摘要

Urban solid waste management continues to remain one of the most neglected areas of urban development in India. The present status of solid waste in Delhi is only marginally being managed through 3 sanitary landfill sites, however a significant portion of the solid waste generated remains unattended. Hence, Delhi needs an efficient and integrated solid waste management system, the successful operation of which depends critically on accurate predictions on solid waste generation. The short-term prediction of future municipal solid waste generation rates can facilitate better planning with respect to collection, personnel and truck utilization, transportation to the landfill and final disposal. In addition, long-term forecasting could also be very useful to select appropriate technologies or scheduling landfill sites. Currently, conventional prediction models are in use, which are based on the demographic and socio-economic factors on a per-capita basis. The present study is directed towards developing a tool, which could be used for the estimation of MSW. For this a mathematical model has been proposed, implemented and tested for the prediction of MSW generation in the city of Delhi. This model is based on a dynamic solid waste time-series analysis. Time series analysis is done through the application of seasonal autoregressive and moving average (sARIMA) model. The model thus developed has been verified using real time data. Predictions are well within 10% variation. Thus, case study of Municipal Corporation of Delhi has been successfully used to demonstrate the application of sARIMA model, in identifying the trends embedded in municipal solid waste generation for a municipal corporation over time. It is predicted that Delhi shall produce 13000 MT in the year 2013. Paper also discusses future planning for MSW management with the help of predicted data.