APPLICATION OF CHATGPT AND GOOGLE COLAB TO CREATE AN ALGORITHM FOR GROUPING AND ROUTING OPTIMIZATION FOR DRUGSTORE INSPECTION: A CASE STUDY OF THE PATHUM THANI PROVINCIAL PUBLIC HEALTH OFFICE

Abstract

Evaluation of drugstore inspections is a task under the supervision of the Provincial Health Office to ensure the quality and standards of pharmaceutical services. The current route planning initially relies on experienced personnel familiar with the area. Therefore, improving or enhancing the efficiency of this process is a significant challenge. This study aimed to develop a method by creating an algorithm using ChatGPT and calculating it through Google Colab to perform clustering and route optimization. The performance of the current expert route planning method was compared with 3 computer-based algorithms. Once the routes were optimized, they were calculated through Google Maps to study the distance and travel time required to visit all locations. This research is a developmental evaluation, using geographic coordinates of the current drugstores and computing them with 3 algorithms, comparing them to expert route planning. The study found that the expert method, which clustered 145 drugstores into 41 groups, resulted in a total distance of 2,755.01 km and a total time of 4,962 minutes. When comparing this with the three algorithms, they produced 21 clusters with distances of 1,313 km, 1,340 km, and 1,359 km, respectively, with reduced travel times. This led to a reduction of 50% in both distance and time. For the case of route sequencing with experts in 41 groups, using the three algorithms yielded different results within each cluster, but the total distance and time remained relatively consistent. The development and testing of the three algorithms can be used for clustering or finding travel sequences, offering a potential future tool to enhance the efficiency of route planning for the inspection of drugstore locations or other similar evaluations. It may be applied for planning in many regional, district, or country-level areas, making work more efficient.