Efficient fish transportation in victory farms limited (VF) Homa-bay County using Hungarian method

Abstract

Fish spoilage during transit has been a major issue for tilapia sh production compa nies, leading to signi cant losses and hindering growth while a ecting food security by reducing the availability of high-quality sh. This study applied the Hungarian method to optimize sh transportation in Victory Farms Limited (VF) Homa-Bay County, with speci c objectives to evaluate the shortest, fastest, and most cost e ective routes for distributing tilapia sh products in the Western Kenya region. Victory Farms currently uses a random delivery approach for distributing its sh, which results in ine ciencies and contributes to spoilage. The Hungarian method was selected due to its ability to systematically determine the optimal match between two sets of elements, making it particularly suitable for optimizing transportation routes in a cost-e ective manner. Primary data on distances, travel times, and costs between key distribution points were collected directly from Victory Farms and used to create a cost matrix, which was analyzed using the Hungarian algorithm. The study identi ed an optimal distribution route for all three objectives: shortest distance, fastest transit time, and lowest cost. The optimal route was determined as follows: KLC → Kondele → Nyalenda → Majengo → Mbale → Chavakali → Kakamega →Mumias→Bungoma→Luanda→Siaya→KLC.Thisroutespanned 293.2 kilometers, took 351.9 minutes in transit, and incurred a cost of Ksh. 14,660. These ndings provide valuable insights into streamlining distribution processes, reducing spoilage, and improving operational e ciency. The results demonstrate that Victory Farms Limited can enhance pro tability and growth by adopting the Hungarian method for route optimization. This approach also contributes to food security by ensuring the availability of high-quality sh. It is recommended that Victory Farms Limited regularly implement the Hungarian method for optimizing routes, while future research could explore integrating real-time environmental and tra c data into the model for additional enhancements.