Optimization of Cloud Kitchen and Service Station Locations

Led a strategic project to optimize the distribution and operational efficiency of 25 cloud kitchens and 50 service stations in San Antonio, employing Python,Pandas for data manipulation, Matplotlib and Geopy for geospatial analysis, and Pulp for linear programming. Spearheading the development and execution of a three-phase approach:

1. Data Acquisition and Preprocessing: We automated the extraction of geolocations (latitudes, longitudes) and zip codes from a CSV dataset using Nominatim Geocoding(API), improving address resolution accuracy.
2. Optimization Model Creation: We designed and solved a binary linear programming model to assign cloud kitchens to service stations with minimized distances, ensuring each kitchen serves exactly two stations while adhering to a minimum distance criterion.
3. Visualization and Analysis: Generated insightful plots and tables to visualize kitchen-station assignments and analyzing distance metrics to guide strategic decision-making. This initiative culminated in the development of an efficient allocation of resources, reducing operational costs and improving service delivery timelines, supported by a robust analytical dashboard for real-time decision-making. The project not only showcased our ability to apply advanced analytics and optimization techniques but also my leadership and collaboration in navigating cross-functional teams towards a data-driven solution, ready for implementation.