Forecasting of Forensic Toxicological Cases at the Institute of Forensic Medicine, Police General Hospital Using Time Series Analysis

Main Article Content

Wachirawich Tungtananuwat Parinya Phoemlap Peerumpa Nootcharoen Weeraya Kongpakpien Kridsada Anuntawuttikul Somsong Lawanprasert

Abstract

Forecasting the information in the future is always necessary for work planning in all institutions. The purpose of this research is to construct the most suitable forecasting model for predicting the quantity of forensic toxicological workload at the Institute of Forensic Medicine, Police General Hospital. The data were collected from alcohol analysis cases in corpses, toxicants/drugs of abuse cases in corpses and suspects (whose samples were sent by the police stations) during 2008 to 2013. Using times series analysis for secular trend, seasonal variation, cyclical variation and irregular variation, the data during 2008-2012 were analyzed and the most suitable models/equations were chosen to predict the quantity of work in 2013. The equations were verified using the 2013 data by comparing the observed data with the predicted data obtained from the equations. The results showed that equations obtained from secular trend demonstrating that cubic equations were most fitted to the data of alcohol cases in corpses, toxicants/drugs of abuse cases in corpses and in suspects. Using the equations, it was shown that mean percentage error of the predicted data versus the observed data of alcohol cases in corpses, toxicants/drugs of abuse cases in corpses and in suspects from the police stations were 11%, 16% and 33%, respectively.

Article Details

Section
บทความพิเศษ