Page 312 - Solid Waste Analysis and Minimization a Systems Approach
P. 312

290     SOLID WASTE ESTIMATION AND PREDICTION





          TABLE 16.14      RANKING AND COMPARISON OF AVERAGE WASTE GENERATION PER
          WASTE GROUP PER COMPANY


                                                                       NUMBER OF       MEAN SOLID WASTE
                                                                       COMPANIES       GENERATION
          WASTE GROUP                              ABBREVIATION        ANALYZED        PER COMPANY

          Wood and lumber manufacturers                 WDM                14                1528.6
          Metal manufacturers                           MLM                10                1313.6

          Food manufacturers                            FDM                 8                 784.8
          Chemical and rubber manufacturers             CHM                16                 749.8

          Paper manufacturers and publishers            PPM                14                 726.0
          Transportation equipment manufacturers        TRM                14                 653.5

          Textile and fabric manufacturers              FBM                13                 584.7
          Electronic manufacturers                      ELM                32                 194.5











                    Also, wood and metal manufacturers generate similar average annual solid waste
                 quantities per company at the 95 percent confidence level. The results of this analysis
                 are important to identify waste groups that generate high quantities of solid waste. The
                 P values indicate that the mentioned waste groups generate different quantities of
                 waste. Specifically, wood and metal manufacturers generate the largest quantities of
                 waste per company. To most effectively use tax dollars, environmental regulators may
                 target these waste groups to achieve higher waste reduction benefits per company over
                 that of other waste groups. These paired t-tests quantitatively ranked the waste groups
                 based on the quantities generated, but qualitative aspects should also be considered.
                 For example, electronic equipment manufacturers generate higher levels of toxic waste
                 than wood manufacturers (the largest waste generating group). This dimension should
                 also be considered when targeting groups.
                    After examining the dominant variables that influence solid waste quantities for
                 each waste group, the variables that do not significantly influence solid waste were
                 analyzed. One of the first insignificant variables was the number of days a company is
                 open per year. This was not a significant variable because the companies were all open
                 for approximately the same number of days per year. This variable did little to differ-
                 entiate the companies.
                    ISO 9000 certification did not enter the solid waste prediction equations, but this
                 is not to say this variable had no significant effect on solid waste generation. All
                 companies that were ISO 14001 certified were also ISO 9000 certified. This indi-
                 cates that ISO 9000 may have an effect on waste generation, but ISO 14001 is a
   307   308   309   310   311   312   313   314   315   316   317