Open Access Open Access  Restricted Access Subscription Access

Algorithm for a PLSR Model and Use in Chemical Process. Case Study: A Fixed Bed Catalytic Reactor

Mr. Serny Klaus A, Mr. Ruette Fernando, Mr. Camacho Jose M


An algorithm based on partial least square regression (PLSR) to enable the construction of a robust multilinear regression model for a chemical process (fixed-bed reactor). The proposed algorithm is tested with experimental data from various works given for all of them a deviation below 3%. This model could be used in scaling up processes being of importance in the design field of process engineering.

Full Text:



Burnham A, Viveros R, Macgregor J. Frameworks for latent variable multivariate regression. Journal of chemometrics. 1996; 10: 31-45p.

Wang X, Kruger U, Lennox B. Recursive partial least squares algorithms for monitoring complex industrial processes. Control Engineering Practice, 2003; 11(6): 613–632p.

Shacham M, Brauner N. Considering the precision of experimental data in the construction of optimal regression models. Chemical Engineering and Processing. 1999; 38: 477–486p.

Ferrer A, Aguado D, Vidal-Puig S, et al. PLS: A versatile tool for industrial process improvement and optimization. Appl. Stoch. Models in Bus. and Ind. 2008; 24 (6): 551−567p.

Li B, Morris J, Martin E. Model selection for partial least squares regression. Chemometrics and intelligent laboratory systems. 2002; 64: 79–89p.

Wold S, Ruhe A, Wold H, et al. The Collinearity Problem in Linear Regression. The Partial Least Squares (PLS) Approach to Generalized Inverses. SIAM Journal on Scientific and Statistical Computing, 1984; 5(3): 735–743p.

Geladi P, Kowalski B. Partial least-squares regression: a tutorial. Analytica Chimica Acta, 1986185. 2016; 1–17p.

Aji A, Schildknecht-Szydlowski N, Faraj A. Partial least square modeling for the control of refining processes on mid-distillates by near-infrared spectroscopy. Oil and Gas Science and Technology. 2004; 59(3): 303–321p.

Burnham A, Macgregor J, Viveros R. Latent variable multivariate regression modeling. Chemo-metrics and intelligent laboratory systems. 1999; 48: 167-180p.

Graciano J, Giudici R, Alves R, et al. A simple PLS-based approach for the construction of compact surrogate models. In Computer Aided Chemical Engineering. 2018; 44(2): 421-426p.

Chebbi R. Optimizing reactors selection and sequencing: Minimum cost versus minimum volume. Chinese Journal of Chemical Engineering. 2014; 22(6): 651–656p.

Gavidia C. Regresión por Mínimos Cuadrados Parciales PLS Aplicada a Datos Variedad Valuados [monograph online]. 2016.

Kresta J, Marlin T, Macgregor J, Development of inferential process models using PLS”, Comp.Chem. & Eng. 1994; 18(7): 597−611p.

Joseph P, Vaswani K, Thazhuthaveetil M. Construction and use of linear regression models for processor performance analysis. IEEE. 2006.

Vega–Vilca J, Guzmán J. Regresión PLS y PCA como solución al problema de multicolineali-dad en regresión múltiple. Revista de matemática: Teoría y aplicaciones. 2011; 18(1): 9–20p.

Hoffmann U, Zanier-Szydlowski N. Portability of near infrared spectroscopic calibrations for petrochemical parameters. Journal of Near Infrared Spectroscopy, 1999; 7(1): 33–45p.

Perezgonzalez J. Fisher, Neyman-Pearson or NHST? A tutorial for teaching data testing. Fron-teirs in Phychology, 2015; 6: 1–11p.

Serny K, Ruette F, Camacho J, et al. Inferential Statistics using Partial Least Square Regression (PLSR) for Chemical Processes. Case Study: A Fixed Bed Catalytic Reactor. Journal of Statistics and Mathematical Engineering. 2020; 6(1): 14–23p.

Andersson M. A comparison of nine PLS1 algorithms. Journal of Chemometrics.. 2009; 23(10): 518–529p.

Hand D, Crowder M, Hand D, et al. Taylor & Francis Group. Chapter 4. Regression methods. Practical Longitudinal Data Analysis, 2018.

Stott A, Kanna S, Mandic D, et al. An online NIPALS algorithm for partial least squares. IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP) 2017; 4177–4181p.

Gil J, Romera R. On robust partial least squares (PLS) methods. Journal of Chemometrics, 1998; 12(6): 365–378p.

González J, Peña D, Romera R. A robust partial least squares regression method with applica-tions. Journal of Chemometrics, 2009; 23(2): 78–90p.

Hubert M, Vanden Branden K. Robust methods for partial least squares regression. Journal of Chemometrics, 2003; 17(10): 537–549p.

Polat E, Gunay S. A new robust partial least squares regression method based on a robust and efficient adaptive reweighted estimator of covariance. RevSta Statistical Journal, 2019; 17(4): 449–474p.

Rousseeuw P, Debruyne M, Engelen S, et al. Robustness and outlier detection in chemometrics. Critical Reviews in Analytical Chemistry. 2006; 36(3–4): 221–242p.

Wakelinc I, Macfie H. A robust PLS procedure. Journal of Chemometrics, 1992; 6(4): 189–198p.

Wise B, Ricker N. Recent advances in multivariate statistical process control improving robustness and sensitivity. IFAC Symposium on Advanced Control of Chemical Processes. 1991.

Corma A, Melo F, Sauvanaud L, et al. Light cracked naphtha processing : Controlling chemistry for maximum propylene production. Catalisis Today. 2005; 108: 699–706p.

Bortnovsky O, Sazama P, Wichterlova B. Cracking of pentenes to C 2 – C 4 light olefins over zeolites and zeotypes Role of topology and acid site strength and concentration. Applied Catalysis A: General. 2005; 287: 203–213p.

Wei Y, Liu Z, Wang G, et al. Production of light olefins and aromatic hydrocarbons through catalytic cracking of naphtha at lowered temperature. Studies in Surface Science and Catalysis. 2005: 158: 1223–1230p.

Yoshimura Y, Kijima N, Hayakawa T, et al. Catalytic cracking of naphtha to light olefins. Catalysis Surveys from Asia. 2001; 4(2): 157–167p.

Spano P. Inferencia robusta en modelos no lineales con respuestas faltantes [monograph online]. 2016.


  • There are currently no refbacks.