Teknik statistika yang digunakan untuk menganalisis permasalahan kualitas dan meningkatkan kinerja proses adalah Statistical Process Control (SPC). Pada teknik SPC terdapat alat untuk mengukur proses dalam keadaan terkendali atau tidak secara statistika yaitu bagan kendali. Pada penelitian ini digunakan bagan kendali multivariat untuk mengontrol kualitas lulusan Prodi Statistika FMIPA USK, yaitu menggunakan bagan kendali Multivariate Exponentially Weighted Moving Average (MEWMA) untuk mengontrol rata-rata proses, Multivariate Exponentially Weighted Moving Variance (MEWMV) untuk mengontrol variabilitas proses, serta Analisis Kapabilitas Proses untuk menilai keseluruhan proses. Data yang digunakan adalah data sekunder yaitu data IPK, lama penyusunan tugas akhir, dan lama studi lulusan Prodi Statistika FMIPA USK tahun 2016-2021 sebanyak 122 orang. Berdasarkan hasil penelitian, menggunakan bagan kendali MEWMA didapatkan bahwa ratarata proses pada kualitas lulusan terkendali secara statistika pada fase II. Pemilihan pembobot paling optimal yaitu λ=0,9. Sedangkan untuk penerapan bagan kendali MEWMV didapatkan bahwa variabilitas proses pada kualitas lulusan juga terkendali secara statistika. Pemilihan pembobot paling optimal yaitu 𝛌=0,9 dan ω=0,3. Hasil perhitungan kapabilitas proses, secara multivariat variabel IPK, lama penyusunan tugas akhir, dan lama studi menunjukkan ketiganya tidak kapabel. Kata kunci⎯ bagan kendali, MEWMA, MEWMV, kapabilitas proses.
ABSTRACT ⎯ Statistical technique used to analyze quality problems and improve process performance is Statistical Process Control (SPC). In this study, multivariate control chart is used to control the quality of the graduates of Statistics Study Program, FMIPA USK, namely using Multivariate Exponentially Weighted Moving Average (MEWMA) control chart to control process average, Multivariate Exponentially Weighted Moving Variance (MEWMV) to control process variability, and analysis ProcessCapability to assess the entire process. Data used is secondary data, namely GPA data, duration of thesis preparation, and length of study for graduates of Statistics Study Program, FMIPA USK in 2016-2021 as many as 122 people. Results of the study, using MEWMA control chart, it was found that the average process for quality of graduates was statistically controlled in phase II. Selection of the most optimal weighting is λ=0.9. Meanwhile, for application of the MEWMV control chart, it was found that the process variability in the quality of graduates was also statistically controlled. The selection of the most optimal weights is 𝛌 =0.9 and ω=0.3. The results of the calculation of process capability, multivariately the GPA variable, length of the final project preparation, and length of the study show that all three are not capable.