diff --git a/arkane/isodesmic.py b/arkane/isodesmic.py
index f45872bcb9..763ebdb7d4 100644
--- a/arkane/isodesmic.py
+++ b/arkane/isodesmic.py
@@ -44,7 +44,11 @@
 
 from __future__ import division
 
+import signal
+
+from lpsolve55 import lpsolve, EQ, LE
 import numpy as np
+import pyomo.environ as pyo
 
 from rmgpy.molecule import Molecule
 from rmgpy.quantity import ScalarQuantity
@@ -265,6 +269,112 @@ def __init__(self, target, reference_set, conserve_bonds, conserve_ring_size):
         self.target_constraint, self.constraint_matrix = self.constraints.calculate_constraints()
         self.reference_species = self.constraints.reference_species
 
+    def _find_error_canceling_reaction(self, reference_subset, milp_software='lpsolve'):
+        """
+        Automatically find a valid error canceling reaction given a subset of the available benchmark species. This
+        is done by solving a mixed integer linear programming (MILP) problem similiar to
+        Buerger et al. (https://doi.org/10.1016/j.combustflame.2017.08.013)
+
+        Args:
+            reference_subset (list): A list of indices from self.reference_species that can participate in the reaction
+            milp_software (str, optional): 'lpsolve' (default) or 'pyomo'. lpsolve is usually faster.
+
+        Returns:
+            ErrorCancelingReaction: reaction with the target species (if a valid reaction is found, else `None`)
+        """
+        # Define the constraints based on the provided subset
+        c_matrix = np.take(self.constraint_matrix, reference_subset, axis=0)
+        c_matrix = np.tile(c_matrix, (2, 1))
+        sum_constraints = np.sum(c_matrix, 1, dtype=int)
+        targets = -1*self.target_constraint
+        m = c_matrix.shape[0]
+        n = c_matrix.shape[1]
+        split = int(m/2)
+        solution = None
+
+        if milp_software == 'pyomo':
+            # Setup the MILP problem using pyomo
+            lp_model = pyo.ConcreteModel()
+            lp_model.i = pyo.RangeSet(0, m - 1)
+            lp_model.j = pyo.RangeSet(0, n - 1)
+            lp_model.r = pyo.RangeSet(0, split-1)  # indices before the split correspond to reactants
+            lp_model.p = pyo.RangeSet(split, m - 1)  # indices after the split correspond to products
+            lp_model.v = pyo.Var(lp_model.i, domain=pyo.NonNegativeIntegers)  # The stoich. coef. we are solving for
+            lp_model.c = pyo.Param(lp_model.i, lp_model.j, initialize=lambda _, i, j: c_matrix[i, j])
+            lp_model.s = pyo.Param(lp_model.i, initialize=lambda _, i: sum_constraints[i])
+            lp_model.t = pyo.Param(lp_model.j, initialize=lambda _, j: targets[j])
+
+            def obj_expression(model):
+                return pyo.summation(model.v, model.s, index=model.i)
+
+            lp_model.obj = pyo.Objective(rule=obj_expression)
+
+            def constraint_rule(model, j):
+                return sum(model.v[i] * model.c[i, j] for i in model.r) - \
+                       sum(model.v[i] * model.c[i, j] for i in model.p) == model.t[j]
+
+            lp_model.constraints = pyo.Constraint(lp_model.j, rule=constraint_rule)
+
+            # Solve the MILP problem using the CBC MILP solver (https://www.coin-or.org/Cbc/)
+            opt = pyo.SolverFactory('glpk')
+            results = opt.solve(lp_model)
+
+            # Return None if a valid reaction is not found
+            if results.solver.status != pyo.SolverStatus.ok:
+                return None
+
+            # Extract the solution and find the species with non-zero stoichiometric coefficients
+            solution = lp_model.v.extract_values().values()
+
+        elif milp_software == 'lpsolve':
+            # Save the current signal handler
+            sig = signal.getsignal(signal.SIGINT)
+
+            # Setup the MILP problem using lpsolve
+            lp = lpsolve('make_lp', 0, m)
+            lpsolve('set_verbose', lp, 2)  # Reduce the logging from lpsolve
+            lpsolve('set_obj_fn', lp, sum_constraints)
+            lpsolve('set_minim', lp)
+
+            for j in range(n):
+                lpsolve('add_constraint', lp, np.concatenate((c_matrix[:split, j], -1*c_matrix[split:, j])), EQ,
+                        targets[j])
+
+            lpsolve('add_constraint', lp, np.ones(m), LE, 20)  # Use at most 20 species (including replicates)
+            lpsolve('set_timeout', lp, 5)  # Move on if lpsolve can't find a solution quickly
+
+            # Constrain v_i to be 4 or less
+            for i in range(m):
+                lpsolve('set_upbo', lp, i, 4)
+
+            # All v_i must be integers
+            lpsolve('set_int', lp, [True]*m)
+
+            status = lpsolve('solve', lp)
+
+            # Reset signal handling since lpsolve changed it
+            try:
+                signal.signal(signal.SIGINT, sig)
+            except ValueError:
+                # This is not being run in the main thread, so we cannot reset signal
+                pass
+
+            if status != 0:
+                return None
+
+            else:
+                _, solution = lpsolve('get_solution', lp)[:2]
+
+        reaction = ErrorCancelingReaction(self.target, dict())
+        for index, v in enumerate(solution):
+            if v > 0:
+                if index < split:
+                    reaction.species.update({self.reference_species[reference_subset[index]]: -v})
+                else:
+                    reaction.species.update({self.reference_species[reference_subset[index % split]]: v})
+
+        return reaction
+
 
 class IsodesmicScheme(ErrorCancelingScheme):
     """
diff --git a/arkane/isodesmicTest.py b/arkane/isodesmicTest.py
index afbcfa374d..29949ab62c 100644
--- a/arkane/isodesmicTest.py
+++ b/arkane/isodesmicTest.py
@@ -185,7 +185,9 @@ class TestErrorCancelingScheme(unittest.TestCase):
 
     def setUp(self):
         self.propene = ErrorCancelingSpecies(Molecule(SMILES='CC=C'), (100, 'kJ/mol'), 'test')
+        self.propane = ErrorCancelingSpecies(Molecule(SMILES='CCC'), (75, 'kJ/mol'), 'test')
         self.butane = ErrorCancelingSpecies(Molecule(SMILES='CCCC'), (150, 'kJ/mol'), 'test')
+        self.butene = ErrorCancelingSpecies(Molecule(SMILES='C=CCC'), (175, 'kJ/mol'), 'test')
         self.benzene = ErrorCancelingSpecies(Molecule(SMILES='c1ccccc1'), (-50, 'kJ/mol'), 'test')
         self.caffeine = ErrorCancelingSpecies(Molecule(SMILES='CN1C=NC2=C1C(=O)N(C(=O)N2C)C'), (300, 'kJ/mol'), 'test')
         self.ethyne = ErrorCancelingSpecies(Molecule(SMILES='C#C'), (200, 'kJ/mol'), 'test')
@@ -199,6 +201,23 @@ def test_creating_error_canceling_schemes(self):
 
         self.assertEqual(isodesmic_scheme.reference_species, [self.butane, self.benzene])
 
+    def test_find_error_canceling_reaction(self):
+        """
+        Test that the MILP problem can be solved to find a single isodesmic reaction
+        """
+        scheme = IsodesmicScheme(self.propene, [self.propane, self.butane, self.butene, self.caffeine, self.ethyne])
+
+        # Note that caffeine and ethyne will not be allowed, so for the full set the indices are [0, 1, 2]
+        rxn = scheme._find_error_canceling_reaction([0, 1, 2], milp_software='lpsolve')
+        self.assertEqual(rxn.species[self.butane], -1)
+        self.assertEqual(rxn.species[self.propane], 1)
+        self.assertEqual(rxn.species[self.butene], 1)
+
+        rxn = scheme._find_error_canceling_reaction([0, 1, 2], milp_software='pyomo')
+        self.assertEqual(rxn.species[self.butane], -1)
+        self.assertEqual(rxn.species[self.propane], 1)
+        self.assertEqual(rxn.species[self.butene], 1)
+
 
 if __name__ == '__main__':
     unittest.main(testRunner=unittest.TextTestRunner(verbosity=2))
diff --git a/environment_linux.yml b/environment_linux.yml
index d499d6fa0b..15f2a9b3e3 100644
--- a/environment_linux.yml
+++ b/environment_linux.yml
@@ -10,11 +10,13 @@ dependencies:
   - cairo
   - cairocffi
   - cantera >=2.3.0
+  - coincbc
   - coolprop
   - coverage
   - cython >=0.25.2
   - dde
   - ffmpeg
+  - glpk
   - gprof2dot
   - graphviz
   - jinja2
@@ -37,6 +39,7 @@ dependencies:
   - pydqed >=1.0.0
   - pygpu
   - pymongo
+  - pyomo
   - pyparsing
   - pyrdl
   - python >=2.7
diff --git a/environment_mac.yml b/environment_mac.yml
index 86dd9c4faa..00060f05a2 100644
--- a/environment_mac.yml
+++ b/environment_mac.yml
@@ -10,11 +10,13 @@ dependencies:
   - cairo
   - cairocffi
   - cantera >=2.3.0
+  - coincbc
   - coolprop
   - coverage
   - cython >=0.25.2
   - dde
   - ffmpeg
+  - glpk
   - gprof2dot
   - graphviz
   - jinja2
@@ -37,6 +39,7 @@ dependencies:
   - pydqed >=1.0.0
   - pygpu
   - pymongo
+  - pyomo
   - pyparsing
   - pyrdl
   - python >=2.7
diff --git a/environment_windows.yml b/environment_windows.yml
index 507ead32c8..6fe391bbe8 100644
--- a/environment_windows.yml
+++ b/environment_windows.yml
@@ -10,11 +10,13 @@ dependencies:
   - cairo
   - cairocffi
   - cantera >=2.3.0
+  - coincbc
   - coolprop
   - coverage
   - cython >=0.25.2
   - dde
   - ffmpeg
+  - glpk
   - gprof2dot
   - graphviz
   - jinja2
@@ -38,6 +40,7 @@ dependencies:
   - pydqed >=1.0.0
   - pygpu
   - pymongo
+  - pyomo
   - pyparsing
   - pyrdl
   - python >=2.7
diff --git a/requirements.txt b/requirements.txt
index aff4494247..b7667d1b8b 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -15,6 +15,8 @@ MarkupSafe # this makes Jinja2 faster
 Jinja2 # this is for rendering the HTML output files
 cairocffi
 yaml
+pyomo # For MILP problems such as Clar structure generation and automatic generation of error canceling reactions
+coincbc # MILP solver
 
 # For postprocessing the profiling data
 argparse