1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 60 61 62 63 64 65 Application of Best First Search Method to Search Nearest Business Partner Location (Case Study: PT Coca Cola Amatil Indonesia, Bandar Lampung) 1 st Heni Sulistiani Faculty of Engineering and Computer Science Universitas Teknokrat Indonesia Bandar Lampung, Indonesia Faculty of Mathematics and Natural Science University of Lampung henisulistiani@teknokrat.ac.id 2 nd Febri Wardani Faculty of Engineering and Computer Science Universitas Teknokrat Indonesia Bandar Lampung, Indonesia febri.wardani@ccamatil.com 3 rd Ari Sulistyawati Faculty of Engineering and Computer Science Universitas Teknokrat Indonesia Bandar Lampung, Indonesia ari.sulistyawati@gmail.com Abstract— PT Coca-Cola Amatil Indonesia is a company that is one of the leading soft drink producers and distributors in Indonesia. PT Coca-Cola Amatil Indonesia has business partners spread in various locations, especially in Bandar Lampung. Every day the goods are delivered by a deliveryman. During the process of shipping goods, Deliveryman is only provided with a piece of paper containing the address and name of the business partner. This makes it difficult for deliverymen who do not know the location of the targeted business partners. In addition, it will also require a lot of time and costs for fuel. Therefore a solution is needed to find the shortest route to reach the location of the intended business partner. In this study, the best first search method was proposed to find the location of business partners with the shortest route. Keywords— algorithm, location, route, search, shortest I. INTRODUCTION PT Coca Cola Amatil is a company that is one of the leading soft drink producers and distributors in Indonesia. The marketing office and the supply of products are spread across all provinces in Indonesia, one of which isi in Lampung Province. PT Coca-Cola Amatil Indonesia has business partners spread across various locations in Bandar Lampung. Shipping goods process is done every day by a deliveryman. Problems that occur in the process of shipping goods include the marketing area of PT Coca Cola Amatil Indonesia Bandar Lampung Sales Office, many outlets that are partners of the company, and deliverymen do not know the route in the area of Bandar Lampung. Thus resulting in length of time and increase costs to search location of business partners. Therefore a solution is needed to find out the shortest route to reach the location of business partners to be visited. There are many algorithms that can be used to determine the shortest path, including Dijkstra's algorithm [1][2][3], Best- First Search algorithm [4][5][6] and A Star (A*) algorithm [1][7][6]. Each algorithm has different ways of working in finding the optimal solution. Dijkstra algorithm works by using a cost function. Each node on the map will be checked and recorded, from the initial node to the final node until the most optimal path is found. The dijkstra algorithm must examine all nodes so that the performance of the algorithm is less effective for finding optimal routes on maps with many routes. While the best first search algorithm works using heuristic functions without calculating costs. The function is able to prioritize checking consecutive nodes and in the right direction. By utilizing the Best-First Search algorithm, the shortest route between the business partners that will be addressed will be known so that it can assist the deliveryman. Therefore in this study it was proposed to apply the best first search method in finding the shortest route. The purpose of this research is to facilitate deliveryman in distributing goods to business partners. This paper is organized as follows. Section 2 describes theory background. Section 3 describes research method. Section 4 describes result and analysis. Section 5 concludes this study. II. THEORY BACKGROUND A. Best First Search Algorithm Search problems are often encountered by many researchers in the field of artificial intelligence. The search method is said to be important because each state describes the steps to solve the problem. There are many search methods that have been used by researchers including hill climbing [8][9][10], dynamic programming [11][12][13], A*[1][7][6], Djikstra [1][2][3], Best First Search [4][5][6], depth first search [14][15], etc. One of the frequently used search methods is best first search. Best First Search is a method that generates a node from the previous node. Best first search selects a new node that has the smallest cost among all leaf nodes that have been raised. Determining the best node is done by using a function called the evaluation function f(n), best-first search evaluation function can be an estimated cost from a node to a goal or a combination of actual costs and estimated costs. Decision Tree of Best First Search Algorithm can be see on figure 1. Proc. ICOMITEE 2019, October 16th-17th 2019, Jember, Indonesia 978-1-7281-3436-9/19/$31.00 ©2019 IEEE 102 ICOMITEE 2019 1570571625 1