Two of the most important supply chain activities are the planning and improvement of
the packing and distribution of products. These can be studied by means of Vehicle
Routing Problems (VRPs) and Packing Problems (PPs). Analyzing these operations
separately, and considering only economic aspects, may result in impractical solutions
and in overlooking the environmental and social aspects of distribution activities. With
the aim of addressing these issues, this thesis presents optimization models and
computational solution procedures for the Vehicle Routing Problem with Loading
Constraints (VRPLC), considering economic, environmental and social aspects (i.e. the
Triple-Bottom-Line (TBL) objectives for sustainability). It also analyzes the effect of
these aspects on routing and packing decisions.
The thesis starts with a systematic review of the PP, VRP and VRPLC literature, to
identify commonly used objective functions and constraints. These are then linked to the
three TBL dimensions. This is followed by the formulation of a VRPLC model, which
considers optimization criteria and operational constraints that have not been considered
simultaneously in previous studies. Although the model can be used to solve problems of
small size, large-sized instances can be computationally intractable and require the use of
an alternative solution method. In this regard, an efficient hybrid heuristic procedure,
which combines a Greedy Randomized Adaptive Search Procedure and a Clarke and
Wright Savings Algorithm, is developed to solve the initial model.
To specify a more relevant VRPLC application, which is efficient not only in economic
terms but in environmental and social criteria as well, these initial findings are then used
in the formulation of a VRPLC+TBL model. This extends the VRPLC model to consider
the TBL dimensions in terms of profits, fuel consumption, safety features for driving on
roads, and equitable distribution of the total payload across the vehicle fleet. Assigning
the same payload to each driver, however, may not be possible and worker dissatisfaction
may arise due to unfair workload assignments. Using the concepts of Simheuristics, the
VRPLC+TBL model is integrated with the design of incentive contracts, to provide a
mechanism to compensate workload imbalances and to further extend the social
dimension.
Overall, this thesis makes a significant contribution to the body of knowledge of routing
and loading operations by providing comprehensive models and efficient computational
solution procedures that are easy to implement. It also addresses some of the gaps
identified in the area of sustainable supply chain management related to the development
of quantitative models, heuristic methods, and the simultaneous consideration of all three
TBL dimensions. Finally, the thesis proposes some interesting ideas for further research.