Pengembangan Portal Penjualan Lobster berbasis Web menggunakan Metode Scrum dengan Fitur Prediksi Harga menggunakan Regresi Linier Berganda

Abstract

This research aims to reveal effective design and content strategies to enhance the attractiveness of a website-based lobster export sales portal. These strategies are developed to capture attention and maintain buyer interest while integrating a one-month forward price prediction feature using linear regression. The research methodology involves implementing the Scrum method for website project management and multiple linear regression to forecast lobster prices. The focus lies on Scrum implementation in designing the sales portal and incorporating lobster price prediction features on the website. The research results demonstrate the successful development of the website using the Scrum framework within three sprints. Feasibility testing with Blackbox Testing aligns with expectations. Multiple linear regression is utilized to predict lobster prices, yielding a forecasted price of 587.3 for the 57th month. With a changing formula each month, the model aims for greater accuracy in anticipating lobster price fluctuations related to seasonal dynamics and specific events. This research provides insights into the implementation of Scrum in website design and practical applications of lobster price prediction in the context of export sales.