Summary of Common Models in Mathematical Modeling
Optimization
Linear programming, semidefinite programming, geometric programming, nonlinear programming, integer programming, multi-objective programming (hierarchical sequence method) , dynamic programming, storage theory, surrogate model, response surface analysis, column generation algorithm
predictive model
Differential equations, wavelet analysis, regression analysis, gray forecasting , Markov forecasting, time series analysis (AR MAMA.RMA ARTMA LSTM neural network), chaos model time series forecasting, support vector machines, neural network forecasting (a lot of machine learning parts coincide)
dynamic model
Differential equation model (ODE SDE DDE DAE PDE) , difference equation model, cellular automata, queuing theory, Monte Carlo stochastic simulation
graph theory model
Shortest path, minimum spanning tree, minimum cost maximum flow, assignment problem, traveling salesman problem, VRPTW path planning, network flow, path planning algorithm (Dijkstra Floyd A D RRT* LPA* D* lite )
evaluation model
Analytic hierarchy process, entropy method , optimal weighting method, principal component analysis , principal component regression evaluation, factor analysis, fuzzy comprehensive evaluation, TOPSIs method , data envelopment analysis, rank sum ratio method, gray comprehensive evaluation method , minimum Square subjective and objective consistent weighting evaluation model, BP neural network comprehensive evaluation method
Statistical analysis model:
Distribution test, mean T test, analysis of variance, analysis of covariance, correlation analysis, chi-square test, rank sum test, regression analysis, Logistic regression, cluster analysis, discriminant analysis, association analysis (Apriori algorithm)
Modern intelligent algorithm:
(extreme value, multi-objective programming, TSP, workshop scheduling, etc.) Simulated annealing, genetic algorithm, particle swarm algorithm, tabu search, immune algorithm, fish swarm algorithm, neural network, ant colony algorithm
other algorithms
Dichotomy method, direct search method, variable range search, single factor optimization method 0.618 method (golden section method), Lagrange multiplier method, trust region algorithm, Euler method\improved Euler method, Newton-Raphson algorithm ( Newton iteration method), quasi-Newton method, gradient descent method
- Classification Problems : KNN, Logistic Regression, Decision Trees, Random Forest,
ADABOOST, GBDTXGBoostlLightGBM, Support Vector Machines, Naive Bayes, Neural Networks - Regression problems : linear regression, LAsSo regression, Ha regression, decision tree regression, regression methods in ensemble learning, support vector regression, Gaussian mixture model, neural network
- Clustering problems : K-means clustering, DSCAN clustering, EM algorithm