Entropy-based k shortest-path routing for motorcycles: A simulated case study in Jakarta

Traffic congestion is a serious problem in rapidly developing urban areas like Jakarta, Indonesia's capital city. To avoid the congestion, motorcycles assisted with navigation apps are popular solution. However, the existing navigation apps do not take into account traffic data. This paper proposes an open-source navigation app for motorcycle by taking into account the traffic data and wide road to avoid congestion. The propose navigation app uses entropy-balanced k shortest paths (EBkSP) algorithm to suggest different routes to different users to prevent further congestion. Tests show that the proposed route planning system in the app gives routes that are significantly shorter than motorcycle routes planned by Google Maps. The EBkSP algorithm also distributes vehicles more evenly among routes than the random kSP algorithm and does so in a practical amount of computing time.