Using Demographics To Predict Future Waste

How can you plan for solid waste disposal when you don't know how much waste your community generates?

Unfortunately, few communities have the resources, much less the finances, to study waste generation. As a result, solid waste officials commonly use national averages, which do not take into account the distinct features of a community.

Historically, researchers have turned to community demographics to predict waste generation. However, little work has been done to determine which variables are most influential.

Demographic variables to consider when predicting generation of waste include median family income, household size and persons per household as well as less conventional indicators such as home lot size, number of households with air conditioners and even the community's annual average temperature.

The University of Illinois at Chicago, with funding from the Office of Solid Waste Research at the University of Illinois Urbana-Champaign, has developed a method to define waste generation within a comprehensive framework of demographic variables.

This framework helps researchers assess the influences of demographic variables on solid waste generation in a more systematic way. The method, developed at the county level, uses demographic and waste generation data from Illinois counties.

The Method To reduce the original number of demographic variables, the research team used data analysis techniques and selected subsets. By using subsets, the researchers were able to describe the major differences between the counties. The variables that did not help describe the differences between counties were then dropped.

For example, if all the counties in a group had similar average housing costs, then the researchers deleted that variable since it had no distinguishing characteristics. On the other hand, if the median age of each county varied, then the variable would be kept in the database.

The following variables remained in the study: the percent of adults educated above the high school level (EDLHIGH); percent of county land covered by farmland (INFARM); median age (MEDAGE); median number of persons per household (MEDHDEN); percent of the total county population employed (PEREMP); and the percent of county population living in urban areas (URBPOP).

To complete the comprehensive framework, these variables were then combined with the amount of solid waste generated by each county.

Municipal solid waste (MSW), which included residential, commercial or institutional refuse, and total solid waste (TSW), which included industrial waste in addition to MSW, were added to the demographic variables.

Researchers found demographic variables EDLHIGH, MEDAGE, PEREMP and URBPOP to influence the MSW and TSW in Illinois.

To double-check the results, the researchers used cluster analysis, which processes county location, topography and other distinguishing characteristics, to classify the counties into 10 demographically similar groups.

When the mean MSW and TSW generation rates of each group were compared to the other groups, a noticeable difference was found in the generation rates due to demographics.

For example, if all 10 groups were ranked from highest to lowest URB-POP, and the solid waste generation rates were then compared, the results would show an overall increase in MSW and TSW generation rates as URBPOP increases (see chart on page 60).

Using The Models The comprehensive framework was used to develop preliminary models. To do this, four other demographic variables that influence the counties' MSW and TSW were used including households per county (HSEHLDS); median income per household (MEDINC); median number of housing units per residential building (MEDUNIT); and population density (PDTOT).

These variable were also used to develop models for the subset waste generation categories of residential (RES), commercial/institutional (C/I) and industrial (IND) waste. The following variables were used for the models:






Note that PEREMP, which is an example of the county's business climate, is included in both the C/I and IND models, but not the RES model.

On the other hand, median income per household appears to influence RES but not the C/I or IND classifications.

Mixed collection routes often leave solid waste officials with the difficult task of distinguishing between waste streams; HSEHLDS may appear in both the RES and C/I models because of this.

The Application Results from the study are being used to create an MSW database which holds basic socio-economic data for each county in Illinois.

With this information, regulatory authorities can compare the estimated solid waste generation of individual counties to others in the same classification group to improve the state's quality control.

County authorities in the same class can use this information to:

* plan jointly;

* perform surveys; or

* share information.

Increasing cooperation among demographically similar counties can prevent the potential use of inappropriate national and state references. In addition, allied counties can design more rational and economical solid waste stream studies to achieve more accurate results for future planning.

Demographic methods can be applied at any geopolitical level, including zip code, town, city, county, state or country.

Using demographics gives solid waste authorities an effective tool to help develop solid waste plans for the future. Next, researchers will have to study the effects of demographic trends on solid waste generation.