Syllabus

Statistical Analysis in Food Processing [8 hrs]
• Principles of inference: population parameters, estimation, and sampling distributions
• Confidence intervals and hypothesis testing
• Type I and Type II errors
• t-tests for two population means (independent and paired samples)
• One-way ANOVA using F-test
• Variance analysis and confidence intervals for variances

Linear Regression [8 hrs]
• Simple Linear Regression (SLR): model assumptions, estimation
• Correlation, coefficient of determination (R²), and hypothesis testing
• Multiple Linear Regression (MLR): matrix formulation, interpretation, and ANOVA
• Model adequacy, multicollinearity

Empirical Model Development [4 hrs]
• Factorial, fractional factorial and rotatable central composite experimental design.
• Developing empirical equations using experimental data.
• Model fitting, adequacy checking, and result interpretation

Artificial Neural Network (ANN) [8 hrs]
• Overview of neural networks and logistic regression
• Cost function, gradient descent, and computation graphs
• Vectorization and implementation of logistic regression
• Forward and backpropagation in neural networks
• Activation functions and training deep neural networks

Optimization using Genetic Algorithm [4 hrs]
• Overview of the biological basis for genetic algorithms.
• Fitness function and binary encoding
• Population formation, selection, crossover, and mutation
• Use of penalty functions and integration with empirical/ANN models

Convolutional Neural Network (CNN) [8 hrs]
• Introduction to Computer Vision
• CNN operations: convolution, pooling, and filters
• Classical architectures: LeNet-5, AlexNet, VGG-16
• ResNet, Inception Network, MobileNet, EfficientNet
• Transfer learning, data augmentation, and implementation tools

To enable students to apply statistical analysis techniques for interpreting and analyzing food process data using hypothesis testing and inference methods.

To equip students with skills to develop predictive models using linear regression, multiple regression, and empirical modeling techniques for food processing applications.

To train students in optimizing food processing parameters using Artificial Neural Networks (ANNs) and Genetic Algorithms.

To help students utilize deep learning models by developing and evaluating Convolutional Neural Networks (CNNs) for food process simulations and predictions.

Understand and apply statistical inference techniques relevant to food process data.

Develop and analyze linear and multiple regression models for predicting food process outcomes.

Design empirical models using appropriate experimental design techniques.

Implement and evaluate ANN models and genetic algorithms for food process optimization.

Develop and evaluate the CNN based models.

R.J. Freund and W.J. Wilson, Statistical Methods, Academic Press , 2nd Edition

Douglas C. Montgomery, Design and Analysis of Experiments, John Wiley & Sons , 5th Edition

H. Das, Food Processing Operations Analysis, Asian Books Private Limited , 1st Edition

M.H. Kutner, C.J. Nachtsheim, J. Neter, W. Li, Applied Linear Statistical Models, McGraw-Hill , 5th Edition

Journal: Trends in Food Science & Technology; Publisher: Elsevier B.V. ; ISSN: 1879-3053

Conference: IFT FIRST Annual Event & Expo, Institute of Food Technologists, USA