Diagram Kendali Nonparametrik Sintetis Exponentially Weighted Moving Average dalam Memantau Rata-Rata Proses dan Penerapannya pada Data Pengukuran Core 4st PT. ABCD
Abstract
Abstract. Control chart is a tools in statistical process control (SPC) used to monitor whether the the production process is stable or not. Control chrat was first developed by Shewhart. However, the Swehart type chart is not sensitive in detecting small shifts and the alternative is the EWMA control chart which is capable of detecting small shifts, but this chart assumming the observation process comes from normal distribution. The altervative is nonparametric EWMA Sign control chart that is appropriate for non normal data. To increase its power, Haq et al. (2018) proposed a synthetic chart that combines EWMA chart and conforming run length (CRL) sub-chart, called SynEWMA chart. This chart has better performance compared to control charts in general. This research applied the SynEWMA nonparametric control diagram to the core 4st data. Core 4st is one of the components in the motor starter dynamo and starter coil whose function is to strengthen the magnetic field and act as a fastener for the motor frame mount. In the phase I, it used non-parametric Shewhart control chart, it was obtained the mean process is µ0 = 485150. Using the SynEWMA control chart in phase II and L=10, a value of k = 1,4679, the control limits obtained were as follows: UCL= 0.8380 and LCL=0.7854. There were several points outside of the control limits or nonconforming. From these points, by comparing them with the L value, it can be concluded that there were out of control processes in observations 2, 3, 4, 5, 18, and 19.
Abstrak. Diagram kendali merupakan suatu alat dalam pengendalian proses statistik (SPC) yang digunakan untuk memantau apakah suatu proses produksi stabil atau tidak. Diagram kendali pertama kali dikembangkan oleh Shewhart. Namun diagram kendali tipe Swehart kurang sensitif dalam mendeteksi pergeseran kecil dan alternatifnya adalah peta kendali EWMA yang mampu mendeteksi pergeseran kecil, namun diagram kendali ini mengasumsikan proses pengamatan berasal dari distribusi normal. Alternatifnya adalah diagram kendali Tanda EWMA nonparametrik yang sesuai untuk data non normal. Untuk meningkatkan kekuatannya, Haq et al. (2018) mengusulkan diagram kendali sintetis yang menggabungkan diagram kendali EWMA dan sub-grafik run length (CRL) yang sesuai, yang disebut diagram kendali SynEWMA. Grafik ini mempunyai kinerja yang lebih baik dibandingkan dengan diagram kendali pada umumnya. Penelitian ini menerapkan diagram kendali nonparametrik SynEWMA pada data core 4st. Core 4st merupakan salah satu komponen pada dinamo starter motor dan koil starter yang fungsinya untuk memperkuat medan magnet dan berfungsi sebagai pengikat dudukan rangka motor. Pada tahap I menggunakan diagram kendali Shewhart non parametrik diperoleh rata-rata proses µ = 4,8515. Dengan menggunakan diagram kendali SynEWMA tahap II dan L=10 diperoleh nilai k = 1,4679 batas kendali diperoleh sebagai berikut: BKA=0.8380 dan BKB=0.7854. Terdapat beberapa titik di luar batas kendali atau tidak sesuai. Dari poin-poin tersebut, jika dibandingkan dengan nilai L maka dapat disimpulkan bahwa terjadi proses diluar kendali pada pengamatan ke-2, 3, 4, 5, 18, dan 19.
References
Abid, M., Nazir, H. Z., Riaz, M., & Lin, Z. (2016). Use of RankedSet Sampling in Nonparametric Control Charts. Journal of the Chinese Institute of Engineers, 3839(April). https://doi.org/10.1080/02533839.2016.1152165
Arnold, Taylor B.; Emerson, John W. (2011). "Nonparametric Goodness-of-Fit Tests for Discrete Null Distributions" (PDF). The R Journal. 3 (2): 34[Dash]39. doi:10.32614/rj-2011-016.
Bourke, P. D. (2018). Detecting a Shift in Fraction Nonconforming Using Run-Length Control Charts with 100 % Inspection. Journal of Qualify Technology, 4065(July). https://doi.org/10.1080/00224065.1991.11979328
Calzada, M.E. and Scariano, S.M. (2001). The Robustness of the synthetic control chart to non-normality. Journal of commun, statist.-simula, 30:2, 311-326.
Crowder SV. A simple method for studying run length distributions of exponentially weighted moving average charts. Dent Tech. 1987; 29: 401- 407.
Crowder SV. Design of exponentially weighted moving average schemes. J Qual Technol. 1989; 21(3): 155- 162.
Goedhart, R., Schoonhoven, M., & Does, R. J. (2019). Nonaprametric control of the conditional performance in statistical process monitoring. Journal Of Quality Technology.
Graham, M. A., Chakraborti, S., & Human, S. W. (2011). A nonparametric EWMA sign chart for location based on individual measurements. Quality Engineering, 23(3), 227–241. https://doi.org/10.1080/08982112.2011.575745
Graham, M.A., Human, S.W., Chakraborti, S. 2009. A Phase I Nonparametric Shewhart-Type Control Chart based on the Median. Journal of Applied Statistics, Vol. 37, No.11, 1795-1813.
Haq, A. (2018). A new nonparametric synthetic EWMA control chart for monitoring process rata-rata. Communications in Statistics: Simulation and Computation, 48(6), 1665–1676. https://doi.org/10.1080/03610918.2017.1422750
Jones-Farmer, L. Allison, Victoria Jordan, and Charles W. Champ. 2009. “Distribution-Free Phase I Control Charts for Subgroup Location.” Journal of Quality Technology 41(3): 304–16.
Lucas JM, Saccucci MS. Exponentially weighted moving average control schemes: properties and enhancements. Dent Tech. 1990; 32(1): 1- 12.
Montgomery, D. C. (2009). Introduction to Statistical Quality Control, Sixth Edition. John Wiley & Sons, New York.
Quraisy, A. (2020). Data Normality Using Kolmogorov-Smirnov and Shapiro-Wilk Tests. J-HEST Journal of Health Education Economics Science and Technology, 3(1), 7–11.
EWMA Control Chart for Process Monitoring.” Revista Colombiana de Estadística 39(2): 167–84. http://dx.doi.org/10.15446/rce.v39n2.58914.
Santoso, B. (2018, Februari 27). Apa Fungsi Yoke dan Pole Core Pada Motor Stater? Diambil kembali dari Pinhome: https://www.pinhome.id/pinhome-home-service/insight/apa-fungsi-yoke-dan-pole-core-pada-motor-starter/#:~:text=Pole%20core%20%3A%20Berfungsi%20sebagai%20penopang,dialiri%20arus%20listrik%20dari%20baterai
Saida, L. (2011). Analisis penerapan bagan pengendali jumlah kumulatif untuk mendeteksi pergeseran rata-rata proses.
Scariano, M. S., & Calzada, M. E. (2009). The Generalized Synthetic Chart. Taylorand Francis
Walpole, Ronald E., Raymond H Myers.; “Ilmu Peluang Dan Statistika untuk Insinyur dan Ilmuawan”, edisi ke-4, Penerbit ITB, Bandung, 1995.
Wu, Z., & Spedding, T. A. (2018). A Synthetic Control Chart for Detecting Small Shifts in the Process Mean. Journal of Quality Technology, 4065(January). https://doi.org/10.1080/00224065.2000.11979969
Yang, Su Fen, Jheng Sian Lin, and Smiley W. Cheng. 2011. “A New Nonparametric EWMA Sign Control Chart.” Expert Systems with Applications 38(5): 6239–43. http://dx.doi.org/10.1016/j.eswa.2010.11.044
