The world faced a major disruption in the form of the coronavirus disease (COVID-19) pandemic, which caused many countries to impose severe restrictions on movement, popularly known as "lockdown." These lockdowns impacted transportation adversely, leading to massive disruptions in global and local supply chains. As the local markets were shut down, more people started turning to e-commerce logistics platforms offering doorstep deliveries of essential items (food and medicines). This resulted in an explosion in demand for such services, and businesses struggled to complete their deliveries. Additionally, the volume of real-time text data suddenly increased, as these customers started sharing their feedback on social media platforms. The availability of real-time raw text data and its popularity for solving complex business problems motivated the development of the approach proposed herein to address last-mile delivery issues.Thus, this paper suggests the use of Twitter data to identify the various grievances of customers about e-commerce logistics platforms. Natural language processing, a popular tool for text analytics, is employed to extract consumer tweets from the Twitter profiles of such businesses and subsequently to clean, process, and analyse them. Issues are categorized and used as objectives in a multiobjective fuzzy vehicle routing problem (VRP). An integrated hybrid fuzzy VRP is developed and coded to solve last-mile delivery issues. Experimental results and comparative analyses highlight the benefits of the novel approach. Managerial insights and scope for future research assist in the further development of the idea.
Real-life challenges require proactive measures. Good transportation services and greener alternatives demand steadfast research in the area. Here, an attempt has been made to address such a problem. A particular case of VRP, with deliveries split into bags and triangular fuzzy travel times, has been modelled to minimize fuel emissions. The concepts of fuzzy rule-based implication for ranking and for comparing fuzzy numbers with numeric values, an expected value model, have been drawn upon. A discrete fuzzyhybridized GA has been developed. Multiple experiments on data in existing works, parameter tuning, and comparative analysis have been performed, thereby corroborating the model's efficacy.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.