In this paper we use ARIMA time series models and statistical process control techniques to determine if changes occurred in the process that generated a series of observations. The procedure is applied to eight crime time series for the city of Houston, Texas: total crime, murder, rape, robbery, aggravated assault, burglary, theft, and motor vehicle theft. The objective of our analysis is to determine if, after the institution of an overtime programme by the Houston Police Department, the different series show some statistical evidence of a change. Our results show that the categories of total crime, robbery, burglary, theft, and motor vehicle theft decreased significantly. Murder and rape remained statistically unchanged, while aggravated assault continued to increase.