calling IBM CPLEX

cplex_demo.py

from forestry import ForestryLinearProgram      # this is project-specific
from PuLP.solvers import CBC            # open-source LinearProgramming library and solver
from creds import BASE_URL, API_KEY     # these are used to id with IBM
from docloud.job import JobClient

my_problem = ForestryLinearProgram('forestry_test.xlsx')    # input file contains all relevant tree data

#my_problem.solve(CBC)     # this is how we used to work, until our problems grew too big for open-source

my_problem.save_to_MPS('forestry_LP.mps')     # here we export to a format that IBM CPLEX can recognize
client = JobClient(BASE_URL, API_KEY)
resp = client.execute(input = 'forestry_LP.mps', output = 'solution.json')

This is a fairly standard linear program/problem in forestry. The Objective Function (1) seeks to maximize Z, which is defined as the total income generated by the decisions in vector x. This vector represents whether stand i is harvested in period j, and is binary (all or nothing). Each stand must be harvested at some point (2). Minimum (3) and maximum (4) volume constraints must be observed. Finally there is an area constraint (5) which we can ignore.

how to output the x values in the json?