الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 180 |  |  |
| | | from | 180 | | |
Abstract
This proposed thesis presents a novel approach in the design of resilient supply chain networks with high ability to resist disruptions. Recently, manufacturers are trying to maximize profits through consolidating facilities (suppliers, factories and distributors) in geographic areas with reduced material, labor and processing costs. Thus, the flow of materials between facilities has been very interconnected over the globe with multiple transportation modes adopted. This has increased the risk of disruptions hitting a certain geographic area and causing a severe damage to proximate facilities. Which may end up in paralyzing the function of the supply chain network and though the depending industries. Also, seeking for reducing supply chain costs, manufacturers are now driven to centralize their functions in facilities with extended capacities. On the disruption of any central facility or transportation line the effect on total system performance could be drastic reflected in high shortage of delivered quantities. This has arouse a question about the configuration of facilities and transportation lines in the supply chain network and its relation with the ability to reduce the impact of disruptions maintaining minimum performance loss. On analyzing the ability of different supply chain network configurations to resist disruptions, researchers have concluded the major effect of three design characteristics:
1) Supply chain density: Denotes the concentration of facilities (suppliers, factories and distributors) in a certain geographic area, where dense supply chain networks are in risk of wide effect disruptions.
2) Node criticality: Denotes the relative importance of any node (facility) within the supply chain network, where a critical node is supposed to have relatively high number of inbound and outbound flows from upstream and downstream nodes. Thus, the effect of critical node disruption on total system performance could be more devastating than the effect of a non-critical node.
3) Flow complexity: Denotes the number of material transporting connections (links) between facilities in the supply chain network. Thus, the disruption of a transportation link in high flow complexity network will result in less shortage of delivered quantities than in a network with less flow complexity.
Realizing the impact of these determinants on supply chain resilience, great attention has been directed to quantify and include them in the supply chain design phase. Previous research efforts focused on measuring and assessing the determinants in existing networks and presented frameworks to include them as design objectives. For this purpose, the proposed thesis presents a multi-period multi-product supply chain mathematical design model. Determinants are first modelled and considered as constraints to the design model with an objective of maximizing profit. Upper or lower limits are considered for each determinant to study its variation on resulting network structures, and hence the ability to resist disruptions. Modelling disruption scenarios with random occurrence and severity has allowed for further assessing the performance of each structure. Outcomes of this phase has allowed for better understanding the direct impact of each determinant on resilience as well as the interaction between determinants. A trade-off is also conducted between induced design resilience and the severity of anticipated disruption to ensure achieving maximum profit. Resilient ranges for each determinant are concluded providing further insight about the selection criteria that should be adopted.
This fruitful break through has allowed for developing the design model to consider maximizing node dispersion and flow complexity as objectives besides maximizing profit. Weighted sum of objectives optimization method is adopted for solving the multi-objective model. where each objective is associated with a weight relevant to its importance in the design optimization with the sum of weights equals to one. The multi-objective model can be used by supply chain designers (architects) in generating resilient and profitable networks based on the set weight for each objective. In order to provide a tool that can be used in generating objectives weights, random disruption scenarios affecting production and transportation capacities were modeled. Assessing the behavior of each network on disruption has allowed for developing weights selection algorithm depending on the form of anticipated disruption and its severity. Sustainable performance on disruption has been maintained for networks generated using the proposed multi-objective design model with weights selection algorithm.
Thesis outcomes present a novel approach in the design of resilient supply chain networks and opens the window for considering further types and forms of disruption. The proposed design tool is adaptive to be applied on supply chain networks with cultural or commercial specific characteristics.