Optimizing Product Delivery through Two-Dimensional Time Warping Demand Allocation under Uncertainty
Prita Meilanitasari (a*), Iwan Vanany (b), Mochamad Nizar Palefi Ma^ady (c), Nisa Isrofi (d)

(a*) Department Logistic Engineering, Universitas Internasional Semen Indonesia, Indonesia
*prita.meilanitasari[at]uisi.ac.id
(b) Department of Industrial and System Engineering, Institut Teknologi Sepuluh Nopember, Indonesia
(c) Department of Information Systems, Institut Teknologi Telkom Surabaya, Indonesia
(d) Department Logistic Engineering, Institut Teknologi Telkom Surabaya, Indonesia

Abstract

In the last three years, the delivery and logistics industry has seen substantial growth, primarily due to the effects of Covid-19. Companies are focusing on minimizing shipping costs by implementing efficient strategies to reduce overall logistics expenses. One such strategy is clustering, which involves grouping regions or requests to transport goods to and from delivery centers. Clustering is essential for customer segmentation and location-based decision-making, as it allows businesses to improve efficiency, customer satisfaction, and profitability. This study has developed an algorithm that streamlines delivery by grouping work areas to make it more efficient and cost-effective. The research used two-dimensional time warping to consider flexible demand locations for goods delivery. The data consisted of two-dimensional delivery point location data. The method consists of three stages: 1) processing data on point distances, 2) clustering using two-dimensional time warping, and 3) validation through silhouette analysis. This research resulted in optimal and efficient demand clustering through location clustering with a Silhouette coefficient value of 0.7 or an accuracy and feasibility level of 70%.

Keywords: Flexible Clustering- Two-Dimensional Time Warping Algorithm- demand uncertainty

Topic: Business and Economic