Impact of population change and unemployment rate on Philadelphia’s waste disposal

Abstract

Predicting waste disposal of a given municipality could be complicated and expensive for government agencies. Lack of a uniform modeling approach, the gap between the scientific community and the government, inaccessibility to the forecasts of variables used in the waste management literature, and budget deficiencies could all result in over-simplification and possibly employing inaccurate modeling approaches for decision makers. This paper portrays the trend of Total Solid Waste (TSW) and Municipal Solid Waste (MSW) disposal of Philadelphia (Pennsylvania, US) with respect to the rate of population change, unemployment rate change, and the current recycling policies. The objective is to develop satisfactory predictive models for the TSW disposal using the same number of variables as currently used by the City of Philadelphia. It is crucial to include an economic factor such as unemployment rate in modeling the waste disposal, especially during economic downturns when economic factors can dominate the effects of population change on waste generation and therefore disposal. Two predictive models are developed using time series analysis and stationary multiple linear regression. The stationary multiple linear regression model yields more accurate predictions for both TSW and MSW disposal of Philadelphia with an approximate level of 8.8% Root Mean Square Percentage Error (RMSPE) and R2 of 0.7. Even the VAR model, with RMSE of 0.15 million tons (RMSPE = 10.7%), provides better estiamtions than does the City of Philadelphia’s current working model.