PERANCANGAN PROTOTIPE SISTEM PENDUKUNG KEPUTUSAN PENENTUAN HARGA SATUAN BARANG PT KERETA API INDONESIA (PERSERO)

• Main Author: Nathaniel Gunawan, Jeremy
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/51358
• Institution: Institut Teknologi Bandung

Unit price is a reference in calculating budget requirements. It needs to be determined accurately to maximize the quality of procurement. In the process of determining the unit price, semi-structured decision making is required because the price obtained for each item of goods is within a certain range. Currently, the determination of unit prices at PT Kereta Api Indonesia (KAI) is carried out for approximately 6000 items of goods using Microsoft Excel Spreadsheet software. The process need be done manually by the user because of incapability of the software to make semi-structured decision. This process is resource intensive and produce low quality and low relevancy results. Therefore, since 2018, PT KAI needs to use assistance from external parties to determine prices, which is not ideal and unsustainable to do in the long term. Decision support system (DSS) is a system that can assist decision making by providing analysis of semi-structured problems. In this study, a DSS prototype are developed to determine the unit price of goods using survey data, historical data on purchases of goods, and data on purchasing areas. The design process of the DSS is carried out by referring to the generalized DSS development process method developed by Marakas and refers to the Turban approach by designing the three components of the DSS subsystem, namely: the model management subsystem, the data management subsystem, and the interface management subsystem. The DSS prototype software was developed using the Microsoft Excel Virtual Basic Application programming language and R. An interface was developed for 2 types of users, namely the material stock function and general work units. There are 6 models that are developed for the DSS, which are: outlier checking model, prediction model, evaluation model, unit price determination model, and price adjustment model for regional locations. The evaluation results showed that the developed DSS succeeded in accurately determining 86.59% of the unit price of goods, and required an algorithm processing duration of 6.7 seconds for each item. These results indicate that the system is feasible to use in real conditions and is proven to improve the quality and efficiency of the unit price determination process at PT KAI.