ABSTRACT
E-commerce companies such as Amazon and Flipkart as well as food delivery aggregators such as Swiggy, Zomato, Uber Eats, etc. are constantly trying to reduce costs during peak delivery times and also during peak demand seasons, such as festival seasons. This is done with a view to minimizing the costs of operations such as fuel costs, vehicle costs, etc. Several computational methods such as heuristics and meta-heuristics have been used to minimize these costs in what can be classified as the vehicle routing problem (VRP). One of the variants that is applicable to the e-commerce industry is the time-dependent vehicle routing problem with time windows (TDVRPTW). This can be solved using meta-heuristics. In this chapter, we illustrate how this NP-hard problem can be solved using the AI technique of genetic algorithm (GA). We have developed a new crossover technique in GA to solve the TDVRPTW which can be used by the e-commerce industry to efficiently plan the routes to minimize the total cost and time and also emissions during deliveries.
