Table 1 Exemplary sustainability indicators and metrics.
Related Figures (7)
4.2.1 Economic and business factors. From an operational and competitive perspective, industries, especially retailers, have seen the quantity of returned products increase significantly. This growth has resulted from increased focus on customer satisfaction, espoused by such corporate practices as total quality management and/or retailers (and other industries) using their liberal return policies as a competitive weapon (Blumberg 1999, Chouinard 2005). The scope of costs and benefits for RL practices ranges from hard costs such as the development Figure 3. A hierarchical diagram of the strategies and strategic metrics for the Strategic Metrics Matrix in figure 2. (See table 2 for abbreviation descriptions.) Figure 4. Activities analysis matrices for local and global RL service providers. term Of analysis if an Open-ended relationship is negotiated) ror cach aiternative. In the Activity Analysis Matrix the vertical axis lists the activities affected by the system under consideration. In this illustrative case, we use five of the six activities for RL identified in literature and summarized in figure 1. The densification activity (which involves the compaction of material) is not relevant to the cell phone case example. Along the horizontal axes, drivers used to determine the exact magnitude of impact on each activity are listed. Driver identification is important to make sure is a convenient method for translating values to a common scale (Keeney and Raiffa 1976). The utility functions section of the matrix is used to determine the conversion. The target row defines a value to which the decision-maker aspires, for example, a value of greater than 0 is desired for NPV. The Upper and Lower rows define the bounds of the values for the utility function calculation as well as what may be viewed as the ‘ideal’ and ‘nadir’ (best and worst) points as viewed by the decision- maker; it is assumed that any va ue beyond the best and worst is a constant utility value (value of one or zero, respectively). The “Type” column of figure 5 specifies the type of utility function is to be used. We assume three basic types of linear utility functions for this methodology ( utility functions as well) and are but can be easily extensible to fuzzy or non-linear expressed as follows: For the peaked utility function the following are expected to hold: where y and z in the peaked utility function are linearly proportional functions. where y and z 1n the peaked utility function are linearly proportional functions. For each metric, an estimated value is either calculated (for metrics which can be calculated easily such as NPV) or developed though scoring or other decision- making techniques for those that cannot be directly calculated (such as Stakeholder Participation and other qualitative metrics). For the examples below, the estimated value is based on giving each alternative a score of 0-5, with 5 the highest. For the Stakeholder Participation metric, the local provider option is given a score of 4, indicating that it performs relatively well on this measure. To facilitate comparison, the estimated value is then converted into a normalized value. For a scored measure such as Stakeholder Participation, the conversion is straightforward. For calculated metrics such as NPV, the conversion is accomplished with the utility function. As can be seen in the figure, the realistic upper limit for NPV was determined to be $400 K. The local provider in this example returns an NPV of $250 K, which is normalized to a value of 3.1. The next step is to apply the weights for each metric that were derived in the Strategic Metrics Matrix. Again taking the NPV example, the normalized value of 3.1 is multiplied by the weight of the metric (0.32) to obtain the score for this metric of 1.0. Likewise, Delivery Performance is calculated as 0.06 x 4 for a score of 0.2. Finally, the scores for all metrics are added to arrive at a total score for this system of 3.7 (rounded from 3.712). The final ctan in the Bwyaluate Altarnatizvec nhace je tn maka a danicann which are calculated in each Strategic Analysis Matrix are shown in this matrix. For this example, the local provider does not do as well on the NPV but does better on most other metrics, making it the preferred alternative based on a weighted total of 3.7 versus 3.4 for the global provider. Table 3. Sensitivity analysis of ultimate decision when varying NPV metric weight. 0.06 (see table 3). This adjustment is necessary so that the summation of the weights is still equal to 1.
