A Forecast Model of Refuse Tonnage With Recapture and Uncertainty Bounds

Abstract

A spreadsheet model for forecasting solid waste tonnages is described. The model uses generation coefficients from the technical literature associated with individual material components (paper, glass, metals, plastics and rubber, organic materials, construction waste, inerts and other) to express the amount of waste produced per capita and per employee in the labor force. Estimates of quantities being generated by two major groups, namely domestic and industrial/commercial/institutional sources are reflected in the model's output. Using this disaggregated format, the model estimates total waste, recaptured waste (as a result of policy initiatives), and net refuse. Confidence intervals are developed using a regression model related to the business cycle. Two worked examples are given for data obtained in Canada.

Modelling commercial/industrial waste generation: a Vienna, Austria case study

“…Figure 10 presents a comparison of the per capita waste data gathered for Vienna in kg employee ‐1 y ‐1 with those presented by McBean & Fortin (1992). The latter data were compiled from a series of studies conducted in Canada between 1988 and 1991, as well as from available references.…”

A review of national municipal solid waste generation assessments in the USA

“…Generation factors, as discussed above, could be created so that they address variations caused by local conditions. The periodic refinement of waste generation coefficients addresses the problem identified by McBean and Fortin (1993) that generation factors change with economic and demographic conditions. However, aggregation schemes, such as for Long Island, or even for the Themelis group, dealing with only 50 states, have not resulted in universal participation, suggesting widespread participation in creating such data may be difficult to ensure.…”

“…Hockett et al (1995) determined, using linear regression, that retail sales and tipping fees were significant predictors of variations in waste generation in North Carolina, but found a great deal of uncertainty-some based on the underlying waste accountings. McBean and Fortin (1993) created a model for Waterloo, Ontario, that used industry and household waste generation rates, and then conducted regression analyses using economic data to determine the effect on annual waste generation rates from economic variability. A model in Illinois, based on multivariate analysis of socioeconomic population descriptors, rankordered solid waste generation by county (Cailas et al, 1996), although it is not clear that meaningful state-level information is generated from rank ordering if the populations of the counties vary much.…”

Estimating construction waste in New Zealand: a focus on urban areas, residential and non-residential building activities

“…Furthermore, McBean and Fortin (1993) utilised a modified version of the per-capita multiplier approach to estimate the total domestic and industrial waste generated annually. On the other hand, some studies argued that the financial value indicator provides a more accurate reflection of construction work.…”