Global Journal of Computer Science and Technology, D: Neural & Artificial Intelligence, Volume 22 Issue 1

Figure 9: c101(100) comparison with [30] (ACOMO) and ACO+K-Means Clustering (PS_KPSO) Figure 10: c101(100) vehicle distribution route Global Journal of Computer Science and Technology Volume XXII Issue I Version I 59 ( )D Year 2022 © 2022 Global Journals Vehicle Routing Problem with Time Window Constrain using KMeans Clustering to Obtain the Closest Customer Comparison Looking at all the results above, it is easily discernible that the ACO+K-Means clustering algorithm has performed way better than the improved Ant Colony algorithm and the normal Ant Colony Algorithm. With lesser number of vehicles employed, lesser carbon emission levels noted and better cost management, the proposed system has shown its effectiveness and viability for usage in the real-world logistics problems. The proposed algorithm PS_KPSO has provided about 10.37%, 46.9%, 61.98% and 78.81% reduction in total costs for 25, 50, 75 and 100 customers while there are about 46.61% , 53.27% and 61.16% reduction in total carbon emissions for 50, 75 and 100 customers, when compared with [30]. Along with the aforementioned improvements, there is 100% customer satisfaction in all the cases. The proposed algorithm (ACO+K-Means Clustering) has outperformed the Modified Ant Colony Algorithm and the original Ant Colony algorithm. Table 3show results compared with Table of that obtained by the proposed algorithm and modified ant colony algorithm. d)