BIRLA INSTITUTE OF TECHNOLOGY AND SCIENCE, Pilani
Pilani Campus
AUGS/ AGSR Division
SECOND SEMESTER 2024-25
COURSE HANDOUT
Date: 04.01.2025
In addition to part I (General Handout for all courses appended to the Time table) this portion gives further
specific details regarding the course.
Course No. : CHE F315
Course Title : Machine Learning for Chemical Engineers
Instructor-in-Charge : AJAYA KUMAR PANI
1. Course Description:
Introduction to machine learning and relevance in Chemical Engineering, Univariate and multivariate
techniques of data processing; Dimensionality reduction; Machine learning techniques for process modeling;
Supervised algorithms (Regression, ANN, SVM etc.); Unsupervised algorithms (PCA, Clustering etc.);
Application to Chemical engineering Problems (reactors, distillation, pumps, heat exchangers etc..) using
suitable computational platforms.
2. Scope and Objective of the Course:
Sophisticated instrumentation coupled with improved data storage facility has made the modern process
industries 'data rich and information poor'. An important step in successful accomplishment of the goals of
Industry 4.0 is effective utilization of this huge resource of plant data in order to achieve reduced downtime,
lowered product rejection and improved process efficiency. This course aims at making the students familiar
with nature of industrial data, data preprocessing, brief theory and application of commonly used
univariate/multivariate statistics and machine learning techniques on industrial data so as to achieve the
following industrial objectives: Adherance to product quality and emission quality norms, detection and
diagnosis of industrial fault.
3. Text Books:
T1: S Sridhar, M Vijayalakshmi. (2021). Machine Learning. OUP..
T2 : S. Dutt, S. Chandramouli, A. K. Das (2021). Applied Machine Learning, 2nd edition. McGraw
Hill.
4. Reference Books:
Kruger, U., & Xie, L. (2012). Statistical monitoring of complex multivariate processes: with applications
in industrial process control. John Wiley & Sons
Ren, J., Shen, W., Man, Y., & DOng, L. (Eds.). (2021). Applications of Artificial Intelligence in Process
Systems Engineering. Elsevier.
Reis, M. S., & Gao, F. (2021). Advanced Process Monitoring for Industry 4.0. Special Issue published in
Processes, MDPI (Free ebook available at [Link]
Christopher M. Bhisop, Pattern Recognition & Machine Learning, Springer, 2006.
Marsland Stephen, Machine Learning – An Algorithmic Perspective, 2e, CRC Press, 2015
1
BIRLA INSTITUTE OF TECHNOLOGY AND SCIENCE, Pilani
Pilani Campus
AUGS/ AGSR Division
5. Course Plan:
Module: Topics to be Covered in Lecture (L) Reference Learning
Lecture No. Sessions Ch./Sec. # Outcome
M1:1 L1: Importance of data science, statistics, Lecture
Introduction artificial intelligence, soft computing and material
machine learning in Chemical
engineering
L2: Nature of industrial data, Gaussianity C – 3 (T1) Understanding
data and data
M2: 2 - 5 L3: what is outlier, effects of outliers. C – 8 (T1) preprocessing
Univariate and Univariate techniques
multivariate 3 edit rule, robust version, box plot
techniques of L4: Multivariate outlier detection
data processing techniques
L5-6: Use of R/MATLAB for outlier
detection of large industrial data
M3: 6-13 L7-9: Correlation analysis, Principal C – 8 (T1) Multivariate
Variable component analysis techniques for
selection, L10: Partial least square analysis dimensionality
feature L11-12: NIPALS and SIMPLS reduction
extraction, algorithm, cross validation, cumulative
dimensionality percentage variance, SCREE plot
reduction L13-14: Use of R and MATLAB for
dimensionality reduction of high
dimensional correlated dataset
M4: 14-19 L14-15: Introduction, common cause and C – 1 (R1) Simple
Statistical special cause variation, Type I and Type techniques of
process II error, process
monitoring Accuracy measurement (FAR, FDR, monitoring
(Univariate MDR, TTD)
techniques) L16: Shewhart control chart
L17: CUSUM control chart
L18: EWMA control chart
L19: Application using MATLAB
M5: 20-26 L20-21: Limitations of univariate Multivariate
Unsupervised techniques, Nature of multivariate data, and machine
machine Hotelling T2 control chart, Multivariate learning
learning EWMA techniques for
techniques L22-23: Use of PCA for multivariate C – 2 (R1) process
process monitoring. monitoring
L-24: Clustering analysis C – 7 (T1)
L25: Use of R and MATLAB
M6: 26-31 L26-27: linear regression, non-linear C – 3 (T1) machine learning
Supervised regression techniques for
machine
2
BIRLA INSTITUTE OF TECHNOLOGY AND SCIENCE, Pilani
Pilani Campus
AUGS/ AGSR Division
learning L28-29: feed forward neural network C – 5 (T1) unknown output
techniques with back propagation prediction
L30: Introduction to support vector C – 4 (T1)
regression
M7: 30-40 L30-31: Distillation process Research Application of
Case studies on L32-33: Fluid catalytic cracking unit articles Techniques to
Industrial L34-35: Simulated CSTR (detection of Process Industry
benchmark data sensor fault, catalyst decay, fouling) data
L36-38: Tennessee Eastman challenge
problem
L39-40: Benchmark simulation model for
biological wastewater treatment process
6. Evaluation Scheme:
Marks Nature of component
Component Duration Date & Time
(%) (Close Book/ Open Book)
Mid-Semester Test 90 Min. 30 (Close Book)
Comprehensive
180 min 40 (Close Book/ Open Book)
Examination
Surprise tests+ 12 (Close Book/ Open Book)
Project (Take home) 18 (Open Book)
+
Toal 4 surprise tests will be conducted during lecture hours (Two before and two after Mid sem).
Best 3 out of 4 will be considered for grading.
7. Chamber Consultation Hour: Tuesday: 5:00 to 6:00 pm. Please email for making a prior appointment.
8. Notices: Will be posted on google classroom
9. Makeup Policy: Make-up is granted only for genuine cases with valid justification and prior permission of
the Instructor-in-charge. No makeup for the surprise tests during lecture hours.
Instructor-in-charge
Course No. CHE F315
