Tan, Chee Kuan (2020) Web-Based Route Optimization System For Logistic Using Agglomerative Clustering And Genetic Algorithm. Final Year Project, UTAR.
Abstract
Route planning is always a difficult task that challenges all the logistics companies. The difficulties of planning a route is because people may not familiar with those addresses, therefore without any help from other existing applications like Google Map or Waze people are not able to find the most suitable route for them to deliver, especially those companies whose have many trucks involved in their daily operation. Besides, there may be a waste of time if the user keeps using Google Maps to check on the location when planning the route. How if there are hundreds of addresses needed to deliver on that particular day, and it is not wise to plan the route manually. Thus, a web-based route optimization is proposed with the objectives of developing a system that is able to find the shortest route to deliver. In this project, evolutionary prototype methodology was selected to smooth the software development according to the scopes and also the requirements. There are three iterations in this project. For the first iteration, use case diagrams and also the preliminary prototype had been developed. In the second iteration, activity diagram, ERD diagram and data flow diagram had been developed. The second iteration continues by the development of the backend logic that communicates with several APIs which are firebase authentication, firestore and google cloud service. The third iteration focuses on finding those missing requirements and enhance the system. To achieve the project’s objectives, several researches on the existing algorithm and also existing similar systems had been done to collect relevant information. In conclusion, this system had been developed with a total period of six months and all of the objectives had been fulfilled. This system enables the logistic company to arrange its route to deliver based on the number of trucks selected by the user. This system will first cluster those addresses based on the number of trucks involved and then perform the route optimization algorithm to suggest the shortest route to the user. Although all of the objectives had been achieved the current system restricted user to make changes on the calculated route directly, the clustering algorithm and the route optimization algorithm applied are not wise enough and several enhancements can be made for this system. To solve the system’s current limitations, several improvements like enable user to set priority on the D/O, enable user to add/remove a particular D/O on the planned route, review on others clustering and optimization algorithm can be performed in the future.
Actions (login required)
![[img]](http://eprints.utar.edu.my/style/images/fileicons/application_pdf.png)
