Document Type : Research Paper


1 Statistical Quality Control and Operations Research Unit, Indian Statistical Institute, 110, Nelson Manickam Road, Chennai- 600029, India.

2 Statistical Quality Control and Operations Research Unit, Indian Statistical Institute, 203 B. T. Road, Kolkata-700108, India.


In the real world, the overall quality of a product is often represented partly by the measured values of some quantitative variables and partly by the observed values of some ordinal variables. The settings for the manufacturing processes of such products are required to be optimized considering the quantitative as well as the ordinal response variables. But the simultaneous optimization of the quantitative and the ordinal response variables are rarely attempted by the researchers. In this paper, a new approach for simultaneous optimization of quantitative and ordinal responses are presented, which are developed by integrating multiple regression techniques, ordinal logistic regression technique, and Taguchi’s Signal-to-Noise Ratio (SNR) concept. The effectiveness of the proposed method is evaluated by analyzing two experimental data sets taken from the literature. The comparison of results reveals that the proposed method leads to the best optimal solution with respect to the total SNR as well as the Mean Square Error (MSE) of individual responses


Main Subjects

[1]      Taguchi, G. (1986). Introduction to quality engineering. Tokyo, Asian Productivity Organization.
[2]      Derringer, G., & Suich, R. (1980). Simultaneous optimization of several response variables. Journal of quality technology12(4), 214-219.
[3]      Logothetis, N., & Haigh, A. (1988). Characterizing and optimizing multi‐response processes by the taguchi method. Quality and reliability engineering international4(2), 159-169.
[4]      Vining, G. G., & Myers, R. H. (1990). Combining Taguchi and response surface philosophies: A dual response approach. Journal of quality technology22(1), 38-45.
[5]      Tsui, K. L. (1999). Robust design optimization for multiple characteristic problems. International Journal of production research37(2), 433-445.
[6]      Wu, F. C. (2004). Optimization of correlated multiple quality characteristics using desirability function. Quality engineering17(1), 119-126.
[7]      Yin, X., He, Z., Niu, Z., & Li, Z. S. (2018). A hybrid intelligent optimization approach to improving quality for serial multistage and multi-response coal preparation production systems. Journal of manufacturing systems47, 199-216.
[8]      Su, C. T., & Tong, L. I. (1997). Multi-response robust design by principal component analysis. Total quality management8(6), 409-416.
[9]      Tong, L. I., & Hsieh, K. L. (2000). A novel means of applying artificial neural networks to optimize multi-response problem. Quality engineering, 13(1), 11–18.
[10]  Tong, L. I., Wang, C. H., & Chen, H. C. (2005). Optimization of multiple responses using principal component analysis and technique for order preference by similarity to ideal solution. The International journal of advanced manufacturing technology27(3-4), 407-414.
[11]  Liao, H. C. (2006). Multi-response optimization using weighted principal component. The international journal of advanced manufacturing technology27(7-8), 720-725.
[12]  Pal, S., & Gauri, S. K. (2010). Multi-response optimization using multiple regression–based weighted signal-to-noise ratio (MRWSN). Quality engineering22(4), 336-350.
[13]  Marcucci, M. (1985). Monitoring multinomial processes. Journal of quality technology17(2), 86-91.
[14]  Franceschini, F., & Romano, D. (1999). Control chart for linguistic variables: A method based on the use of linguistic quantifiers. International journal of production research37(16), 3791-3801.
[15]  Wang, K., & Tsung, F. (2010). Recursive parameter estimation for categorical process control. International journal of production research48(5), 1381-1394.
[16]  Li, J., Tsung, F., & Zou, C. (2014). A simple categorical chart for detecting location shifts with ordinal information. International journal of production research52(2), 550-562.
[17]  Wu, F. C. (2008). Simultaneous optimization of robust design with quantitative and ordinal data. International journal of industrial engineering: Theory, applications and practice15(2), 231-238.
[18]  Karabulut, G. B. (2013). Comparison of methods for robust parameter design of products and processes with an ordered categorical response (Doctoral Dissertation, Middle East Technical University). Retrieved from
[19]  Box, G. (1986). Discussion of testing in industrial experiments with ordered categorical data by VN Nair. Technometrics28, 295-301.
[20]  Hamada, M., & Wu, C. F. J. (1986). Discussion. Technometrics28(4), 302-306.
[21]  Agresti, A. (2010). Analysis of ordinal categorical data (Vol. 656). John Wiley & Sons.
[22]  Nair, V. N. (1986). Testing in industrial experiments with ordered categorical data. Technometrics28(4), 283-291.
[23]  Koch, G. G., Tangen, C., Tudor, G., & Stokes, M. E. (1990). Discussion: Strategies and issues for the analysis of ordered categorical data from multifactor studies in industry. Technometrics32(2), 137-149.
[24]  Chipman, H., & Hamada, M. (1996). Bayesian analysis of ordered categorical data from industrial experiments. Technometrics38(1), 1-10.
[25]  Jeng, Y. C., & Guo, S. M. (1996). Quality improvement for RC06 chip resistor. Quality and reliability engineering international12(6), 439-445.
[26]  Wu, F. C., & Yeh, C. H. (2006). A comparative study on optimization methods for experiments with ordered categorical data. Computers & industrial engineering50(3), 220-232.
[27]  Asiabar, M. H., & Ghomi, S. F. (2006). Analysis of ordered categorical data using expected loss minimization. Quality engineering18(2), 117-121.
[28]  Bashiri, M., Kamranrad, R., & Karimi, H. (2012). Response optimization in ordinal logistic regression using heuristic and meta-heuristic algorithm. Journal of Sharif university28, 79-92.
[29]  Hsieh, K. L., & Tong, L. I. (2001). Optimization of multiple quality responses involving qualitative and quantitative characteristics in IC manufacturing using neural networks. Computers in industry46(1), 1-12.
[30]  León, R. V., Shoemaker, A. C., & Kacker, R. N. (1987). Performance measures independent of adjustment: an explanation and extension of Taguchi’s signal-to-noise ratios. Technometrics29(3), 253-265.
[31]  Phadke, M. S. (1995). Quality engineering using robust design. Prentice Hall PTR.