<![CDATA[Abstract. The availability of clean and suitable for consumption water is one of the efforts to improve the welfare of the community and is the responsibility of the government. Therefore, in order to ensure that the quality of distributed water meets the standards, it is very important to carry out regular water quality control. Control chart are one of the satatistic tools used in statistical quality control that have been useful for practitioners to produce quality goods and services. If there are more than two quality characteristics, control of these variables needs to use a multivariate control chart. In many practical applications, only individual observations are available n = 1. For these conditions, a Multivariate Exponentially Weighted Mean Squared (MEWMS) control chart can be used to control multivariate dispersions. In this study, the MEWMS control diagram was applied to the quality data of the water production process of PDAM Hazora, Jember Regency consisting of 7 water quality variables. The results of the analysis show that the MEWMS control diagram of the actual data, for each of its observational data is within the control limits at the weighting ω = 0.1; 0,5; 0.9 then, it can be concluded that the process is in control. The simulation results are based on mean vector parameters and data covariance matrix in control, showing that the MEWMS control chart that detects the fastest shift is the diagram using ω = 0.1 with the mean vector shift detected at the time of the 61st observation and for its dispersion at the 115th observation. Abstrak. Tersedianya air bersih dan layak konsumsi merupakan salah satu upaya peningkatan kesejahteraan hidup masyarakat dan menjadi tanggung jawab pemerintah. Oleh karena itu, demi memastikan kualitas air yang terdistribusikan telah memenuhi standar, sangat penting untuk melakukan pengendalian kualitas air secara berkala. Diagram kontrol adalah salah satu alat satatistik yang digunakan pada pengendalian kualitas statistik yang telah berguna bagi praktisi untuk menghasilkan barang dan jasa yang berkualitas. Jika terdapat lebih dari dua buah karakteristik kualitas, pengendalian terhadap variabel-variabel tersebut perlu menggunakan diagram kontrol multivariat. Dalam banyak penerapan praktis, hanya pengamatan individual yang tersedia yaitu n sama dengan 1. Untuk kondisi ini, diagram kontrol Multivariate Exponentially Weighted Mean Squared (MEWMS) dapat digunakan untuk mengendalikan dispersi multivariat. Pada penelitian ini diagram kontrol MEWMS diaplikasikan pada data kualitas proses produksi air PDAM Hazora Kabupaten Jember yang terdiri atas 7 variabel kualitas air. Hasil analisis menunjukkan bahwa diagram kontrol MEWMS data aktual, untuk setiap data pengamatannya berada dalam batas-batas kendali pada pembobot]]>
