Bus Stops Location and Bus Route Planning Using Mean Shift Clustering and Ant Colony in West Jakarta

TL;DR

This paper proposes an optimized bus stop placement and route planning in West Jakarta using mean shift clustering and Ant Colony algorithms, aiming to reduce traffic congestion effectively.

Contribution

It introduces a novel combination of mean shift clustering and Ant Colony algorithms for bus stop location and route optimization in Jakarta.

Findings

Bus stop location error rate of 0.07%

Maximum bus stop distance of 350 meters

Route area covering 32 km

Abstract

Traffic Jam has been a daily problem for people in Jakarta which is one of the busiest city in Indonesia up until now. Even though the official government has tried to reduce the impact of traffic issues by developing a new public transportation which takes up a lot of resources and time, it failed to diminish the problem. The actual concern to this problem actually lies in how people move between places in Jakarta where they always using their own vehicle like cars, and motorcycles that fill most of the street in Jakarta. Among much other public transportations that roams the street of Jakarta, Buses is believed to be an efficient transportation that can move many people at once. However, the location of the bus stop is now have moved to the middle of the main road, and it is too far for the nearby residence to access to it. This paper proposes an optimal location of optimal bus stops in West Jakarta that is experimentally proven to have a maximal distance of 350 m. The optimal location is estimated by means of mean shift clustering method while the optimal routes are calculated using Ant Colony algorithm. The bus stops locations rate of error is 0.07% with overall route area of 32 km. Based on our experiments, we believe our proposed bus stop plan can be an interesting alternative to reduce traffic congestion in West Jakarta.