Following the generic framework presented in the previous section, the methodology to quantify and measure the sustainability performance of products [the product sustainability index (ProdSI)] and processes [process sustainability index (ProcSI)] in sustainable manufacturing practices is presented here. The hierarchical approach to identify product and process sustainability metrics incorporating the three key aspects of sustainable manufacturing and the complete sets of product and process sustainability metrics are presented. The process of normalizing, weighting, and aggregating these metrics to compute the ProdSI and ProcSI is also presented.

Hierarchical Approach to ProdSI and ProcSI Development

While generic definitions of sustainable products and sustainable manufacturing provide general guidance in identifying factors or elements that can evaluate product sustainability, their identification and quantitative evaluation for assessing sustainability performance of a specific product or manufacturing process are challenging and complex tasks (Fiksel et al. 1998). This is due to the wide range of aspects to be considered in evaluating product and process sustainability, the difficulties in quantifying many sustainability aspects (especially the social aspects), and the inherently heterogeneous nature of the data needed for sustainability evaluation which makes it difficult to combine them for an overall assessment.
At this point, it is important to make a distinction must be made between performance metrics and indicators. While an indicator provides qualitative or quantitative information about performance of a specific phenomenon, environment, or area, a performance metric provides a quantitative measure that is required for overall product sustainability assessment. The approach presented here considers sustainability performance metrics which must be measurable, relevant and comprehensive, understandable and meaningful, manageable, reliable, accessible, and measurable in a timely manner (Feng et al. 2010).
The proposed ProdSI and ProcSI methodologies have a hierarchical structure that breaks product and process sustainability down to individual metrics through a five- and four-level process, respectively. These levels are index (ProdSI/ProcSI), sub-index, cluster, sub-cluster (for ProdSI only), and individual metric, as presented in Fig. 6.

To address the challenges of defining product sustainability metrics, a top-down approach was followed. This hierarchical approach ensures that the individual metrics are comprehensive and cover all major aspects of product sustainability. The identified sustainability metrics are however generic and can be applied to any type of product or manufacturing process by customization. The five-level hierarchical structure developed can be described as follows:

• ProdSI – the overall aggregated product sustainability performance index
• Sub-index – the three aspects of the TBL: economy, environment, and society
• Cluster – major elements or factors of product sustainability within each of the three TBL categories
• Sub-cluster – decomposition of clusters to more specific aspects of product sustainability
• Individual metric – a quantifiable and measurable attribute or property related to a single parameter or indicator in each sub-cluster that is measured through out the total product life-cycle

The evaluation of the overall ProdSI and ProcSI is done using a bottom-up approach to aggregate the individual metrics to provide an overall product and process sustainability assessment. The ProdSI and ProcSI are calculated through a series of operational steps including data collection for individual metrics measurement and data collection and data normalization, weighting, and aggregation. The product and process sustainability metrics are presented next.

Product Sustainability Metrics

By expanding the six previously identified major product sustainability elements of environmental impact, societal impact, functionality, resource utilization and economy, manufacturability, and recyclability and remanufacturability, a more comprehensive set of 13 clusters was developed for product sustainability evaluation. These clusters are categorized under the three categories of the TBL: economy, environment, and society, as illustrated in Fig. 7.

The complete set of individual product metrics under the economy, society, and environment subclusters is presented in Tables 1–3, respectively. These metrics can be customized for the product being evaluated, considering its functionality and performance. Further, in order to comprehensively evaluate the product sustainability, the measurement of each of these metrics must be made across the four product life-cycle stages (pre-manufacturing, manufacturing, use, and post-use), depending on their applicability.

Process Sustainability Metrics

Following the criteria mentioned before, a comprehensive set of metrics for manufacturing process sustainability assessment was identified. The chosen metrics are categorized under six clusters that represent the process-related elements of sustainable manufacturing: manufacturing cost, energy consumption, waste management, environmental impact, operator safety, and personnel health. A description of the complete set of individual process metrics under each cluster is presented in Tables 4–9. Similarly, these metrics can be customized for a specific manufacturing process.

ProdSI and ProcSI Evaluation Process

The product and process sustainability metrics provide individual measures, but do not directly provide an overall assessment of product or process sustainability. The proposed ProdSI and ProcSI methodologies aggregate the metrics to provide an overall product and process sustainability assessment, respectively. The details of data normalization, weighting, and aggregation methods are presented in the following subsections.

Normalization

Due to the heterogeneous nature of the sustainability metrics, the physical measurements of individual metrics cannot be directly aggregated. Therefore, all the individual metrics must be converted to a single normalized scale. In the ProdSI and ProcSI methodology, the individual metrics are normalized to a single scale from 0 to 10, where 0 represents the worst case and 10 represents the best case. Generally, a score of 0–4 would indicate a “poor” status, “average” with a score of 4–6, “good” with a score of 6–8, and “excellent” with a score of 8–10.
A single standard normalization method that can be applied for all metrics does not exist; the normalization of each individual metric is case specific and depends on several factors including the unit of measure, the limits of the measured value, whether the individual metric is positively or negatively correlated with overall product sustainability, and the existence of benchmarks or standard reference points for normalization. Establishing reference points is essential to normalize the different units of measurement. Benchmarks can be set up based on earlier generations of the same product, standards, regulations, or expert opinions. The normalization can be done using a continuous scale or a discrete scale from 0 to 10.

Weighting and Aggregation

Weightings are assigned for each element in the ProdSI and ProcSI (individual metrics, subclusters, clusters, and subindices) to balance the normalized values based on their relative importance or level of impact. Typically, a higher weighting is assigned to elements with a higher importance or impact level which must be determined considering many aspects such as regional variations in legislation, expert opinions, monetary valuation, and consumer value requirements. Weighting is a very sensitive process, and it affects the accuracy of the sustainability assessment. Therefore, it is important to be objective when assigning weights to the elements when computing the ProdSI and ProcSI.
Currently, there are no universal or standard weighting methods that can accurately capture the relative importance of sustainability metrics or indicators. As such, the most suitable of several weighting methods can be applied. The first is assigning equal weighting, which is considered simple and transparent. However, the shortcoming of equal weighting is that it does not truly reflect the relative importance of the aggregated elements. The second is soliciting experts’ opinions using surveys and questionnaires. Once the surveys and questionnaires are collected, weighting can be assigned by simply averaging the weights assigned by different experts or by following more complex statistical mechanisms such as the analytic hierarchy process (AHP), previously developed for product sustainability evaluation (Gupta et al. 2010).
Once the weights are assigned, the normalized metrics are aggregated to calculate the scores for the subclusters, clusters, subindices, and ProdSI/ProcSI. The aggregation follows a bottom-up approach as presented in (Eqs. 2–4).

Sub-cluster Level

Cluster Level

where
wp is the weighting factor for the sub-index QP
t is the number of sub-index

The normalization and weighting processes are usually associated with subjective judgments (Singh et al. 2012). This affects the sensitivity and accuracy of the product sustainability assessment. Sensitivity analysis and expert evaluations can help reduce the effects of the subjectivity of normalization and weighting and increase the accuracy of the assessment.