The Application of Statistical Process Control Techniques for Quality Improvement of the Municipal Wastewater-Treated Process

Statistical process control techniques are useful tools for monitoring the production process and detecting abnormal process behavior due to special causes. Once the special causes for abnormal process behavior are identified and consequently eliminated, the process can be further improved. The aim of this study is to apply univariate and multivariate statistical process control techniques to enhance the monitoring of a wastewater treatment process and achieve a higher effluent quality. Phase I, Shewhart univariant control charts and Hotelling’s T² multivariate control chart were developed for non-correlated and correlated variables of a wastewater treatment plant, respectively. Five representative water quality parameters: turbidity, total suspended solids (TSS), chemical oxygen demand (COD), total Kjeldahl nitrogen (TKN) and fecal coliform were investigated. The interpretation of the Phase I for both types of control charts initially revealed that the process was out of statistical control conditions for all the investigated variables. T² decomposition technique revealed that the main contributors for the out-of-control points were turbidity with 67% (average T² = 28.86) followed by TKN 25% (average T² = 31.12). The assignable causes for the observed abnormalities were the result of seasonal variations with respect to the temperature in such hot climates. Control charts proved their applicability for the wastewater process as a quick and efficient monitoring strategy despite the complex nature of the wastewater and the contribution of the hot climate in the Arabian Gulf region.