This study presents an enhanced multi-criteria decision-making framework for ranking emergency shelters, addressing the unique needs of displaced persons beyond mere survival. By defining comprehensive criteria in collaboration with the Emergency Management Organization, the model ensures a holistic approach to shelter selection. The study applies these criteria to a case study in Manitoba, Canada, utilizing an improved Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) model integrated with entropy-based methods for precise, unbiased weight calculations. This enhancement increases reliability compared to conventional TOPSIS models. The involvement of field experts bridges the gap between theoretical models and real-world applications, making the decision-making process more practical. Beyond mathematical improvements, this research incorporates the human aspects of displaced populations’ needs, ensuring a balance between quantitative and qualitative factors. The results demonstrate optimized shelter selection, improved prioritization, and more efficient resource allocation, ultimately reducing response times, minimizing disruptions, and preventing repeated relocations during evacuations in crisis situations.
2 August 2021 in International Journal of Public Administration
Understanding the factors that affect bus ridership, particularly at the agency level, is essential for improving urban mobility and sustainability. This study addresses a significant gap in transit literature by proposing a Bayesian method to identify agency-specific factors influencing bus ridership. Using data from the National Transit Database (NTD) of United States from 2007 to 2017, we examine bus ridership across 45 transit agencies in the United States. Our research goes beyond existing studies by exploring agency-specific effects, allowing us to compare how variables such as population, gas prices, and subsidies impact ridership across different systems. We employ fixed-effects and random-effects models, along with a Bayesian framework, to capture variations and identify differences among agencies. This Bayesian approach allows for the incorporation of prior knowledge and uncertainty, enhancing the robustness of our results. By utilizing a select group of relevant variables, our method empowers transit agencies to discern which factors most profoundly influence their ridership patterns, facilitating data-driven decision-making tailored to their specific contexts. Generalizing trends among transit agencies can be challenging due to the significant variations in bus ridership and other influencing factors. This research addresses this issue by identifying the unique challenges faced by individual transit agencies and offering tailored solutions to meet their specific needs.