INCORPORATING ENVIRONMENTAL CONCERNS IN SUPPLY CHAIN OPTIMIZATION - Environmentally Conscious Materials Handling - page 126

responsible activities (such as reduced packaging and joint recycling
efforts) are a function of the extent of shared responsibility within the
supply chain. Vachon and Klassen (2006b) study the operational perfor-
mance of a supply chain operating under a green project partnership ,
involving both upstream and downstream efforts to prevent pollution.
Using data from the packaging industry, they find that downstream (i.e.,
closer to the customer) green project partnerships are positively linked
to quality, flexibility, and environmental performance, while upstream
partnerships are associated with better delivery performance. A later
study by Vachon and Klassen (2006a) finds that collaboration in the
supply chain plays an even more important role as corporations attempt
to gravitate toward environmental sustainability. In line with their
previous study, they find that a firm's collaboration with its suppliers is
closely linked with process-based performance, while collaboration with
customers is associated with product-based performance. With respect
to horizontal collaboration, decisions related to the endorsement of
specific certifications should be collaboratively conducted across possibly
competing firms, given the ambiguities associated with the myriad of
standards and certifications present in industries and the noisy market
signals that accompany them.

5 APPROACHES TO OPTIMIZATION

In general terms, an optimization problem involves a decision (or variable) space
and a criteria (or objective) space. The appropriateness of approaches used to
solve an optimization problem depends on the characteristics of these spaces.
The preceding sections discussed how the three factors — legislative, economic,
and social — entail changes to the decision space and/or the criteria space of an
SCO problem. For example, the traditional SCO objective of solely maximizing
profits fails to incorporate intangible benefits derived from environmental efforts.
Traditional variables (such as product-process combinations) may have to factor
in uncertainties in the evolution of environmental regulation. Thus, recent trends
have created a need for the use of alternative approaches to formulating SCO
problems. In this section, we provide recommendations of optimization meth-
ods that are capable of accommodating the SCO complexities discussed in the
preceding sections.

5.1 Nonlinear Programming (NLP)

Nonlinear dependencies, such as between decisions and outcomes of interest,
are inevitable in practice. For example, emissions abatement costs are typically
convex increasing in the extent of abatement desired. Nonlinear dependencies