我正在尝试使用熟悉谷歌或工具。我尝试了Employee调度python示例的简化版本。
from __future__ import print_function
import sys
from ortools.constraint_solver import pywrapcp
def main():
# Creates the solver.
solver = pywrapcp.Solver("employee_scheduling")
num_nurses = 3
num_shifts = 3
num_days = 1
# [START]
# Create shift variables.
shifts = {}
for j in range(num_nurses):
for i in range(num_days):
shifts[(j, i)] = solver.IntVar(
0, num_shifts - 1, "shifts(%i,%i)" % (j, i))
shifts_flat = [shifts[(j, i)] for j in range(num_nurses)
for i in range(num_days)]
# Create nurse variables.
nurses = {}
for j in range(num_shifts):
for i in range(num_days):
nurses[(j, i)] = solver.IntVar(
0, num_nurses - 1, "shift%d day%d" % (j, i))
# Set relationships between shifts and nurses.
for day in range(num_days):
nurses_for_day = [nurses[(j, day)] for j in range(num_shifts)]
for j in range(num_nurses):
s = shifts[(j, day)]
solver.Add(s.IndexOf(nurses_for_day) == j)
# Create the decision builder.
db = solver.Phase(shifts_flat, solver.CHOOSE_FIRST_UNBOUND,
solver.ASSIGN_MIN_VALUE)
# Create the solution collector.
solution = solver.Assignment()
solution.Add(shifts_flat)
collector = solver.AllSolutionCollector(solution)
solver.Solve(db, [collector])
print("Solutions found:", collector.SolutionCount())
print("Time:", solver.WallTime(), "ms")
print()
if __name__ == "__main__":
main()
正如你所看到的,我唯一保留的限制是轮班和护士之间的关系。在num_nurses=3、num_shifts=3和num_days=1的情况下,解算器能够找到6个解。但是,如果将num_shifts更改为2,则解算器将返回0个解。这不是也应该有三个解决方案吗(指派一名护士,不指派另外两名护士(?
事实证明,这是employee_scheduling目前编写方式的限制。它的重写工作目前正在进行中,应该在几周内完成。
https://github.com/google/or-tools/issues/932
我推荐这个版本的轮班调度,它实现了一些不同的约束:
https://github.com/google/or-tools/blob/master/examples/python/shift_scheduling_sat.py