Uncertainty modelling and risk optimization approach through fuzzy modelling and evolutionary algorithms for post-pandemic supply chain

• Main Author: Evangelista, Danielle Grace D.
• Format: text
• Language: English
• Published: Animo Repository 2021
• Subjects: Business logistics; Manufacturing
• Online Access: https://animorepository.dlsu.edu.ph/etdm_mem/1; https://animorepository.dlsu.edu.ph/context/etdm_mem/article/1002/viewcontent/Evangelista2.pdf
• Institution: De La Salle University

The COVID-19 pandemic has caused a great impact on many industries worldwide such as in supply chain. Its operations and overall performance were threatened due to the effects of such event. This is because lack of knowledge and unpredictability causes uncertainty thereby exposing the supply chain to a risk of many possible outcomes. Hence, uncertainties go together with supply chain risks since the existence of uncertainty can lead to high supply chain risks, and both can cause significant and negative impacts in a company if not managed effectively and efficiently. Though there are several studies abroad on uncertainty modelling and risk management, there are no known published studies yet regarding supply chain modelling considering the effects of COVID-19 in Philippine firms. Since the COVID-19 pandemic is novel, such supply chain model doesn’t exist where robust and stochastic modelling will not be feasible due to their conservative nature and absence of prior data, respectively. Therefore, there is a need to develop a supply chain model by uncertainty modelling to be solved using evolutionary methods such as Particle Swarm Optimization (PSO) and Genetic Algorithm (GA) can be used for risk optimization where rigorous approaches are not suitable. Two (2) respondent companies (A and B) agreed to be respondents for this study to use their knowledge based on their experiences during the pandemic. Quantitative data from the last quarter of 2020 to the first quarter of 2021 was requested and provided by the companies. These were modelled using fuzzy uncertainty modelling with the objectives of minimizing downstream supply chain costs and minimizing travel time of finished goods from the manufacturing plants to the market and then optimized using evolutionary algorithms, Genetic Algorithm and Particle Swarm Optimization which was able to produce equal results. Afterwards, the minimum cost and time calculated by both evolutionary algorithms were used as input for a fuzzy inference engine to determine the satisfaction values. By quantifying decisionmaker satisfaction, this can be used as a useful tool in the decision-making process to ensure satisfactory business operations and be prepared for any possible risks brought on by future disruptions. Based on the survey analysis with decisionmakers, the paper was able to point out the difficulties experienced by each company and their performance during the pandemic and has also contributed managerial implications to address such difficulties and to improve supply chain resilience. The study also provided possible approaches in choosing the best solution to accept in terms of feasibility and overall satisfaction in which decisions can be based on. In reference to the data provided and results produced, the study recommends prioritizing practicality, thus the best solutions for Company A and Company B are the minimum cost and minimum traveling time values provided with the highest satisfaction and highest feasibility.