Use of control charts with regression analysis for autocorrelated data in the context of logistic financial budgeting

- Logistics budget variables (autocorrelated data) have not been addressed using control charts.
- The loading unit is a good predictor of logistics budget variables.
- Control charts for scaled residuals are useful to monitor budget variables.
- Three types of control charts for regression scaled residuals are tested.
- Control charts with regression are an effective strategy in this context.

The aim of this paper is to explore whether the use of control charts with regression analysis is an effective way to evaluate financial budget requests (autocorrelated data) in the transport logistics sector. First, the variables are selected. Second, a regression analysis is performed to model the financial variables. Third, three types of traditional control charts are tested (individuals, CUSUM and EWMA), using simulation to monitor the regression scaled residuals. The results show that the individual control chart of 2.7-sigma offers an appropriate performance for the context of this study. This paper provides new evidence regarding a type of variable and context not reported in the literature. In addition, it proposes a control chart approach of scaled regression residuals, with two differentiators: (1) residuals offer better practical interpretation and (2) regressions do not incorporate the time variable, as traditionally occurs, but a missionary process variable (loading units) and a control one.