3MAH · chemiskyy · Mar 30, 2026 · Mar 30, 2026 · Apr 3, 2026 · Apr 3, 2026
diff --git a/CMakeLists.txt b/CMakeLists.txt
@@ -350,7 +350,7 @@ if(WIN32 AND BUILD_PYTHON_BINDINGS)
 endif()
 
 # Compile public executables (with main functions)
-set(EXECUTABLES solver identification L_eff Elastic_props ODF PDF)
+set(EXECUTABLES solver L_eff Elastic_props ODF PDF)
 foreach(exe_name ${EXECUTABLES})
   add_executable(${exe_name} software/${exe_name}.cpp)
   target_link_libraries(${exe_name} PRIVATE simcoon ${CMAKE_DL_LIBS})

diff --git a/README.md b/README.md
@@ -13,7 +13,7 @@ Simcoon is developed with the aim to be a high-quality scientific library to fac
 Simcoon integrates
 - a easy way to handle geometrical non-linearities : Use of Lagrangian measures, Eulerian measures and cumulative strains considering several spins : Jaumann, Green-Naghdi, Xi-Meyers-Bruhns logarithmic. With this last measure, cumulative strain correspond to a logarithmic strain measure and is the standard measure utilized for our constitutive laws.
 
-Simcoon is a C++ library with emphasis on speed and ease-of-use, that offers a python interface to facilitate its use. Its principle focus is to provide tools to facilitate the implementation of up-to-date constitutive model for materials in Finite Element Analysis Packages. This is done by providing a C++ API to generate user material subroutine based on a library of functions. Also, Simconnn provides tools to analyse the behavior of material, considering loading at the material point level. Such tools include a thermomechanical solver, a software to predict effective properties of composites, and a built-in identification software (using a combined genetic-gradient based algorithm)
+Simcoon is a C++ library with emphasis on speed and ease-of-use, that offers a python interface to facilitate its use. Its principle focus is to provide tools to facilitate the implementation of up-to-date constitutive model for materials in Finite Element Analysis Packages. This is done by providing a C++ API to generate user material subroutine based on a library of functions. Also, Simcoon provides tools to analyse the behavior of material, considering loading at the material point level. Such tools include a thermomechanical solver and a software to predict effective properties of composites. Parameter identification can be performed using Python with scipy.optimize (e.g. differential_evolution) and the simcoon Parameter/Constant key system
 
 Simcoon is mainly developed by faculty and researchers from University of Bordeaux and the I2M Laboratory (Institut de d'Ingénierie et de Mécanique). Fruitful contribution came from the LEM3 laboratory in Metz, France, TU Bergakademie Freiberg in Germany and the TIMC-IMAG laboratory in Grenoble, France. It is released under the GNU General Public License: GPL, version 3.
 

diff --git a/docs/cpp_api/index.rst b/docs/cpp_api/index.rst
@@ -9,7 +9,7 @@ Simcoon's C++ library provides high-performance implementations for:
 - **Continuum Mechanics**: Tensor operations, strain/stress transformations, kinematics
 - **Constitutive Models**: Hyperelastic, viscoelastic, and viscoplastic material models
 - **Homogenization**: Mean-field homogenization methods for composite materials
-- **Simulation Tools**: Solvers, identification algorithms, and utilities
+- **Simulation Tools**: Solvers and utilities
 
 .. note::
    The C++ API is designed for advanced users who need direct access to the 

diff --git a/docs/cpp_api/simulation/identification.rst b/docs/cpp_api/simulation/identification.rst
diff --git a/docs/cpp_api/simulation/index.rst b/docs/cpp_api/simulation/index.rst
@@ -2,8 +2,8 @@
 Simulation
 ==========
 
-The Simulation module provides tools for running simulations, including solvers,
-identification algorithms, and mathematical utilities.
+The Simulation module provides tools for running simulations, including solvers
+and mathematical utilities.
 
 .. toctree::
    :maxdepth: 2
@@ -12,7 +12,6 @@ identification algorithms, and mathematical utilities.
    maths
    rotation
    solver
-   identification
    phase
 
 Overview
@@ -23,7 +22,6 @@ This module contains simulation and numerical tools:
 - **Maths**: Mathematical utilities (random numbers, statistics, solvers)
 - **Rotation**: Comprehensive 3D rotation tools with ``Rotation`` class and Voigt notation support
 - **Solver**: Material point simulation solvers
-- **Identification**: Parameter identification algorithms
 - **Phase**: Phase management and properties
 
 What's New in Simcoon 2.0

diff --git a/docs/cpp_api/simulation_overview.md b/docs/cpp_api/simulation_overview.md
@@ -2,7 +2,7 @@
 
 ## Introduction
 
-The Simulation module provides the infrastructure for executing multi-physics simulations, parameter identification, and optimization. It includes the solver framework, phase management, mathematical utilities, and identification algorithms for material parameter calibration.
+The Simulation module provides the infrastructure for executing multi-physics simulations. It includes the solver framework, phase management, and mathematical utilities.
 
 ## Module Organization
 
@@ -88,104 +88,9 @@ Results management:
 - **read.hpp** - Phase file parsing
 - **write.hpp** - State serialization
 
-### 3. **Identification** - Parameter Identification Framework
+### 3. **Maths** - Mathematical Utilities
 
-Inverse analysis tools for material parameter calibration from experimental data.
-
-#### **Optimization Framework:**
-
-##### **identification.hpp**
-Main identification driver supporting multiple optimization algorithms:
-- Genetic Algorithms (GA)
-- Gradient-based methods (Levenberg-Marquardt)
-- Hybrid strategies
-
-##### **Core Components:**
-
-**parameters.hpp**
-Defines optimization variables:
-```cpp
-class parameters {
-  public:
-    double value;           // Current value
-    double min_value;       // Lower bound
-    double max_value;       // Upper bound
-    string key;             // Parameter identifier
-    int ninit;              // Number of initializations
-};
-```
-
-**constants.hpp**
-Fixed values during optimization:
-```cpp
-class constants {
-  public:
-    double value;
-    string key;
-    int ninit;
-};
-```
-
-**individual.hpp**
-Solution candidate in population-based methods:
-- Parameter vector
-- Cost function value
-- Constraint violations
-- Fitness ranking
-
-**generation.hpp**
-Population management for evolutionary algorithms:
-- Population initialization
-- Selection operators
-- Crossover and mutation
-- Elitism strategies
-
-**methods.hpp**
-Optimization algorithm implementations:
-- Cost function evaluation
-- Gradient computation (numerical/analytical)
-- Hessian approximation
-
-**optimize.hpp**
-High-level optimization loop:
-- Convergence checking
-- Iteration management
-- History tracking
-- Checkpointing
-
-**opti_data.hpp**
-Experimental data management:
-- Loading experimental files
-- Data interpolation
-- Weight assignment
-
-**doe.hpp** (Design of Experiments)
-Sampling strategies for parameter space exploration:
-- Latin Hypercube Sampling (LHS)
-- Random sampling
-
-**script.hpp**
-Script interpretation for identification workflows.
-
-#### **Cost Function Definition:**
-
-The identification minimizes a weighted sum of squared residuals:
-
-\f[
-f(\mathbf{p}) = \sum_{i=1}^{N_{exp}} \sum_{j=1}^{N_{pts}} w_{ij} \left( y^{exp}_{ij} - y^{sim}_{ij}(\mathbf{p}) \right)^2
-\f]
-
-where:
-- \f$ \mathbf{p} \f$ = parameter vector
-- \f$ N_{exp} \f$ = number of experimental datasets
-- \f$ N_{pts} \f$ = number of data points per dataset
-- \f$ w_{ij} \f$ = weight for point j in experiment i
-- \f$ y^{exp}_{ij} \f$ = experimental observation
-- \f$ y^{sim}_{ij}(\mathbf{p}) \f$ = simulation prediction
-
-### 4. **Maths** - Mathematical Utilities
-
-Mathematical tools supporting simulation and identification:
+Mathematical tools supporting simulation:
 
 #### **rotation.hpp**
 Rotation operations for objective stress integration:
@@ -241,7 +146,7 @@ Random number generation:
 - Seeding control
 - Reproducibility support
 
-### 5. **Geometry** - Geometric Primitives
+### 4. **Geometry** - Geometric Primitives
 
 Geometric representations for multi-phase materials:
 
@@ -312,42 +217,6 @@ solver(umat_name, props, nstatev, psi, theta, phi);
 
 Results are written to output files specified in the control file.
 
-## Parameter Identification Workflow
-
-### 1. Define Parameters to Identify
-
-```cpp
-vector<parameters> params;
-params.push_back({"E", 50000, 100000, "Young"});
-params.push_back({"sigma_Y", 100, 500, "Yield"});
-```
-
-### 2. Load Experimental Data
-
-```cpp
-opti_data data;
-data.import("tensile_test.txt");
-data.weight = 1.0;
-```
-
-### 3. Configure Optimization
-
-```cpp
-int method = 0;  // Genetic Algorithm
-int maxiter = 100;
-double tolerance = 1e-6;
-```
-
-### 4. Run Identification
-
-```cpp
-identification(method, params, constants, data_files);
-```
-
-### 5. Retrieve Optimal Parameters
-
-The identified parameters are written to `parameters_results.txt`.
-
 ## Advanced Features
 
 ### Multi-scale Modeling
@@ -375,14 +244,6 @@ The solver automatically adjusts time step size based on:
 
 Some operations support parallel execution (using OpenMP):
 - Multiple phase response
-- Population-based optimization (outdated)
-
-### Checkpointing
-
-Identification can be checkpointed for:
-- Recovery from interruptions
-- Continuation of long-running optimizations
-- Sensitivity analysis
 
 ## Configuration Files
 
@@ -448,20 +309,13 @@ Block_1:
 
 ## References
 
-1. **Genetic Algorithms:**
-   - Goldberg, D. E. (1989). *Genetic Algorithms in Search, Optimization, and Machine Learning*. Addison-Wesley.
-
-2. **Micromechanics and Multi-scale Modeling:**
+1. **Micromechanics and Multi-scale Modeling:**
    - Qu, J., & Cherkaoui, M. (2006). *Fundamentals of Micromechanics of Solids*. Wiley.
 
-3. **Parameter Identification:**
-   - Mahnken, R., & Stein, E. (1996). "Parameter identification for viscoplastic models based on analytical derivatives of a least-squares functional and stability investigations." *International Journal of Plasticity*.
-
-4. **Fischer-Burmeister Method:**
+2. **Fischer-Burmeister Method:**
    - Fischer, A. (1997). "Solution of monotone complementarity problems with locally Lipschitzian functions." *Mathematical Programming*.
 
 ## See Also
 
 - [Continuum Mechanics Module](continuum_mechanics_overview.md) - Material models and constitutive functions
 - [Solver API](solver/) - Detailed solver documentation
-- [Identification API](identification/) - Parameter identification reference
diff --git a/docs/index.rst b/docs/index.rst
@@ -13,7 +13,7 @@ Simcoon is a free, open-source C++ library for simulating multiphysics systems.
 [microgen](https://github.com/3MAH/microgen) for CAD and meshing of heterogeneous materials and
 [fedoo](https://github.com/3MAH/fedoo), our finite-element solver, the project provides a comprehensive toolset for analyzing heterogeneous materials.
 
-Simcoon aims to be a high-quality scientific library for analysing complex, nonlinear system behaviour. It emphasizes performance and ease of use and exposes a Python interface to simplify workflows. The principal focus is to provide a C++ API to generate user-material subroutines for finite-element packages. Simcoon also includes tools to analyse material-point behaviour under loading, such as a thermomechanical solver, a tool to predict effective properties of composites, and a built-in identification tool that combines genetic and gradient-based algorithms.
+Simcoon aims to be a high-quality scientific library for analysing complex, nonlinear system behaviour. It emphasizes performance and ease of use and exposes a Python interface to simplify workflows. The principal focus is to provide a C++ API to generate user-material subroutines for finite-element packages. Simcoon also includes tools to analyse material-point behaviour under loading, such as a thermomechanical solver and a tool to predict effective properties of composites. Parameter identification can be performed using Python with scipy.optimize (e.g. differential_evolution) and the simcoon Parameter/Constant key system.
 
 Simcoon supports geometric nonlinearity using Lagrangian and Eulerian measures, and cumulative strains with several objective rates (Jaumann, Green--Naghdi, and Xi--Meyers--Brühns logarithmic). The logarithmic cumulative-strain measure is the default used by the library's constitutive laws.