Prediction, Visualization, and Optimization of Resources Using Time-Series Forecasting Models and Simplex Linear Programming

Abstract

Crime is one of the major problems of countries all over the world, and the Philippines is no exception. Crime prediction and prevention strategies are vital for police forces to face inevitable increases in the crime rate as a side effect of the growth of the urban population. This paper focuses on the prediction of crime rates. It also focuses on the development and testing of the effectiveness of the optimization model in reducing the crime rate score reduction considering the number of mobility using Simplex Linear Programming and regression analysis. Various time-series forecasting models were applied in the crime dataset using the SAS tool. Datasets were extracted from fourteen (14) municipal police stations of Rizal Province, which contains historical data of crime statistics from 2013 to 2017 and mobility resources for each Police station. MAPE was used to determine the accuracy of each model. The prediction results can be useful for the police stations to identify problematic regions to patrol and the predicted values for mobility derived from the optimization model can be a valuable information in decision making specifically in the disposition of mobility for a given locality to suppress crime so that law and order can be maintained properly and there is a sense of safety and well-being among the citizens in the province.