Scenario Creation

In addition to model-based parameter estimation, parmest can create scenarios for use in optimization under uncertainty. To do this, one first creates an Estimator object, then a ScenarioCreator object, which has methods to add ParmestScen scenario objects to a ScenarioSet object, which can write them to a csv file or output them via an iterator method.

This example is in the semibatch subdirectory of the examples directory in the file scenario_example.py. It creates a csv file with scenarios that correspond one-to-one with the experiments used as input data. It also creates a few scenarios using the bootstrap methods and outputs prints the scenarios to the screen, accessing them via the ScensItator a print

# ____________________________________________________________________________________
#
# Pyomo: Python Optimization Modeling Objects
# Copyright (c) 2008-2026 National Technology and Engineering Solutions of Sandia, LLC
# Under the terms of Contract DE-NA0003525 with National Technology and Engineering
# Solutions of Sandia, LLC, the U.S. Government retains certain rights in this
# software.  This software is distributed under the 3-clause BSD License.
# ____________________________________________________________________________________

import json
from os.path import join, abspath, dirname
import pyomo.contrib.parmest.parmest as parmest
from pyomo.contrib.parmest.examples.semibatch.semibatch import SemiBatchExperiment
import pyomo.contrib.parmest.scenariocreator as sc


def main():

    # Data: list of dictionaries
    data = []
    file_dirname = dirname(abspath(str(__file__)))
    for exp_num in range(10):
        fname = join(file_dirname, 'exp' + str(exp_num + 1) + '.out')
        with open(fname, 'r') as infile:
            d = json.load(infile)
            data.append(d)

    # Create an experiment list
    exp_list = []
    for i in range(len(data)):
        exp_list.append(SemiBatchExperiment(data[i]))

    # View one model
    # exp0_model = exp_list[0].get_labeled_model()
    # exp0_model.pprint()

    pest = parmest.Estimator(exp_list)

    scenmaker = sc.ScenarioCreator(pest, "ipopt")

    # Make one scenario per experiment and write to a csv file
    output_file = "scenarios.csv"
    experimentscens = sc.ScenarioSet("Experiments")
    scenmaker.ScenariosFromExperiments(experimentscens)
    experimentscens.write_csv(output_file)

    # Use the bootstrap to make 3 scenarios and print
    bootscens = sc.ScenarioSet("Bootstrap")
    scenmaker.ScenariosFromBootstrap(bootscens, 3)
    for s in bootscens.ScensIterator():
        print("{}, {}".format(s.name, s.probability))
        for n, v in s.ThetaVals.items():
            print("   {}={}".format(n, v))


if __name__ == "__main__":
    main()

Note

This example may produce an error message if your version of Ipopt is not based on a good linear solver.