Integrating Sustainability into a Goal Programming Exercise

ABSTRACTThis paper discusses a sustainability exercise for use in a management science course. Specifically, we discuss an exercise using goal programming and Excel Solver for making supplier selection decisions incorporating a triple bottom line approach (economic, environmental and social performance objectives). The multiple, conflicting objectives and the qualitative nature of the social performance objective require the use of multi-criteria decision-making. Our goal programming exercise requires only Excel and could be expanded to include additional triple bottom line criteria.JEL: C6, M11KEYWORDS: Sustainability, Management Science, Curriculum, Triple Bottom Line, Goal Programming(ProQuest: ... denotes formulae omitted.)INTRODUCTIONOur Supply Chain Management (SCM) program at UW Oshkosh started integrating sustainability into our major in the Fall Semester of 2006 and continues to integrate sustainability into all of our SCM courses. The first widespread definition of sustainability was presented in Our Common Future (World Commission on Economic Development, 1987, p. 8) in which sustainable development was defined as "development that meets the needs of the present without compromising the ability of future generations to meet their own needs." Other researchers (e.g., Elkington (1994, 1998)) expanded the definition of sustainability to include the triple bottom line criteria of economic, environmental, and social performance. The least understood and under-researched of the three bottom lines is social performance. Mass and Bouma (as cited in Castro & Chousa, 2006) divided the social performance criteria into two broad categories: internal measures (education, training, safety, health care, employee retention, and job satisfaction) and external measures (sponsoring, volunteer work, investment in society, and stakeholder involvement). Norman and MacDonald (2004) argued that it is impossible to calculate a social performance bottom line in the same way that an income statement is created. Summing a company's performance on various social performance measures into a single bottom line is problematic due to: (a) the question of what units to use to express social performance, and (b) the manner in which social performance often is expressed-using percentages, which cannot be added or subtracted into a single meaningful measure. However, even though managers cannot calculate a bottom line for social performance, we argue that managers still could make value judgments and comparisons concerning which social performance criteria are more important. Multi-criteria decision-making (MCDM) methods, and in particular, goal programming, work well for making these value judgments and comparisons.Goal programming is an extension of linear programming in which the objective function measures the minimization of unwanted deviations from goals (targets). As discussed by Romero (2004), two of the most common types of objective functions for goal programming models are lexicographic and weighted. The lexicographic type of achievement function, used later in our paper, leads to preemptive, or prioritized, goals. As described by Anderson, Sweeney, Williams, Camm, and Martin (2012), goal programming problems with preemptive priorities are solved by finding the solutions for a sequence of linear programming models with different objective functions: Priority Level 1 goals are considered first, Priority Level 2 goals second, etc. At each step of the solution procedure, a revision in the solution is allowed only if it causes no reduction in the achievement of higher priority goals previously minimized. Anderson et al. (2012) discussed two types of constraints in a goal programming model: hard constraints, which are typical linear programming constraints that cannot be violated, and soft constraints, which correspond to goal equations and can be violated but with a penalty for doing so (the penalty is represented by deviation variables). …