Enabling Industrial Ecology through the Forecasting of Durable Goods Disposal: Televisions as an Exemplar Case Study

Canadian Journal of Administrative Sciences, Jun 2004 by Linton, Jonathan D, Yeomans, Julian Scott, Yoogalingam, Reena

Disposal of Televisions

Having considered the lifecycle and likelihood of failure of the television, we will now consider the disposal of the television once it has failed in a fashion that is not economically repairable. Once the television has failed, it is ready for disposal; however, being ready for disposal and disposed of are separate issues. We recognize that many products are stored for a period of time prior to their disposal, the television being one such product (EPA, 1998a; Linton, 1999; Ron & Delpeut, 1996), but there is no available information on this subject. It is proposed that 25-35% of televisions are disposed of in the year they fail and the rest are disposed of within the next 10 years according to a triangular distribution (see Figure 2). The disposal quantity for the remaining years is based upon the number of units remaining and the likelihood of their disposal in a particular year, determined from the use of the triangular distribution.

Usable Waste Stream

The usable waste stream for leaded glass from televisions depends not only on product storage and failure, but also on the weight of the CRTs from these televisions and the reclamation rate for televisions. In previous studies, the weight of the CRT has been estimated as a single fixed value, usually either 40 or 44 pounds (20 kg) (EPA, 1998b; Mizuki & Schuldt, 1998; Ron & Delpeut, 1996). However, the weight of the television tube depends primarily on the size of the viewing screen of the television and, hence, the weight of CRTs varies tremendously. In small televisions, it is only about 10 pounds, while in the largest CRT-based televisions it is about 90 pounds. To take into account the variation in weight between televisions, we have modeled the weight of an individual television as being a random value from the distribution shown in Figure 3. Using the parameters from this distribution of an individual television's CRT weight, together with the Central Limit Theorem and the value determined for the total number of televisions disposed of, we can easily generate a value for the total CRT weight entering the waste flow in any particular year.

Recovery of Waste

Next, we consider the proportion of discarded televisions that is actually reclaimed. Over time, through the development of recycling infrastructure, consumer education, and habit, reclamation programs can capture an increasing proportion of disposed products (EPA, 1998a; Tchobanoglous, Theisen, & Vigil, 1993). To reflect this trend, reclamation has been modeled with a straight-line increase from a maximum of 10% in 2000 (the start of the forecast period) to 90% in 2050 (the end of the forecast period). To incorporate an element of variability into the progressively increasing reclamation rate, the actual reclamation percentage is allowed to randomly vary over a 5% range beneath the projected yearly maximum.

Forecasting Future Television Sales

Because this model addresses disposal behaviour for a long time into the future, it is necessary to consider televisions that have not yet been sold. Consequently, future sales of televisions must be forecast and the effect of competing technologies must be considered. The sales of CRT-based televisions depend upon two factors: the demand for televisions and the displacement (if any) of CRT-based televisions by FP displays or other technologies. One extreme is to assume that CRT-based televisions will not face competition from any other technology over the next 50 years and that the concerns regarding the lead content of the glass will remain. In such a scenario, the important question regarding future television sales is the actual number of televisions sold. Television sales throughout the last decade indicate a stable market penetration rate and relatively flat sales (EIA, 1998). Consequently, for a "no technological change" scenario, sales are modeled each year for the next 50 years as a random value between the highest and lowest sales numbers over the last decade.