# -*- coding: utf-8 -*-
"""
Created on Mon Jan 21 10:33:38 2019
@author: zhangchaoyu
"""
import re
import copy
import os
import math
import datetime
import sys
import wmi
import hashlib
def read_data_csv(f):
#f=open(r'数据.csv')
data = []
f.readline()
while(True):
row = f.readline()
if row == ',\n' or row == '':
break
row = re.split(',', row)
row[1] = float(row[1])
row[2] = float(row[2])
row[4] = float(row[4])
row[5] = float(row[5])
if row[6] != "":
row[7] = float(row[7])
row[8] = float(row[8])
row[9] = int(math.ceil(float(row[9])))
row[10] = int(math.ceil(float(row[10])))
row[11] = float(row[11])
row[12] = float(row[12])
row[13] = float(row[13])
row[14] = float(row[14])
row[15] = float(row[15])
row[16] = float(row[16])
row[17] = float(row[17])
row[18] = float(row[18])
row[19] = float(row[19])
row[20] = float(row[20])
data.append(row)
return data
def read_parameter_csv(f):
#f=open(r'配置文件.csv')
parameter = []
#接货仓人员配置
f.readline()
for i in range(3):
row = f.readline()
row = re.split(',', row)
parameter.append(float(row[1]))
#前置分拣配置
f.readline()
f.readline()
for i in range(3):
row = f.readline()
row = re.split(',', row)
parameter.append(float(row[1]))
#分拣中心配置
f.readline()
f.readline()
row = f.readline()
row = re.split(',', row)
parameter.append(float(row[1]))
#车辆配置
vehicles = []
f.readline()
f.readline()
f.readline()
for i in range(2):
row = f.readline()
row = re.split(',', row)
row[1] = float(row[1])
row[2] = float(row[2])
row[3] = float(row[3])
row[4] = float(row[4])
row[5] = float(row[5])
row[6] = float(row[6])
vehicles.append(row)
parameter.append(vehicles)
#依据单量百分比确定个数
f.readline()
f.readline()
row = f.readline()
row = re.split(',', row)
parameter.append(float(row[1]))
#车辆复用规则
f.readline()
f.readline()
f.readline()
cycle = []
for i in range(3):
row = f.readline()
row = re.split(',', row)
row[0] = float(re.split('-',row[0])[0])
row[1] = float(row[1])
cycle.append(row[:2])
parameter.append(cycle)
f.readline()
f.readline()
f.readline()
warehouses = []
while(True):
row = f.readline()
if row == ',\n' or row == '':
break
row = re.split(',', row)
row[1] = float(row[1])
row[2] = float(row[2])
row[3] = float(row[3])
row[4] = float(row[4])
warehouses.append(row[:5])
parameter.append(warehouses)
return parameter
def data_clean(data, parameter):
warehouses = parameter[10]
ware = []
for w in warehouses:
ware.append(w[0])
data1 = []
for d in data:
if d[0] in ware:
data1.append(d)
return data1
def data_base_read(data, parameter):
#把仓和首分拣收集出来
warehouses = parameter[10]
delvs_first = []
delvs_second = []
for i in range(len(data)):
if data[i][3:6] not in delvs_first:
delvs_first.append(data[i][3:6])
if data[i][6:9] not in delvs_second:
delvs_second.append(data[i][6:9])
warehouses.sort()
delvs_first.sort()
delvs_second.sort()
return [warehouses, delvs_first, delvs_second]
def matrix_zero(m, n):
M = []
for i in range(m):
temp = []
for j in range(n):
temp.append(0)
M.append(temp)
return M
def calculate_Spherical_distance(lat1, lng1, lat2, lng2):
lat1 = lat1*math.pi/180
lng1 = lng1*math.pi/180
lat2 = lat2*math.pi/180
lng2 = lng2*math.pi/180
return 6371.393*math.acos(max(min(math.cos(lat1)*math.cos(lat2)*math.cos(lng1-lng2)+math.sin(lat1)*math.sin(lat2),1),-1))
#return 6371.393*math.acos(math.cos(lat1)*math.cos(lat2)*math.cos(lng1-lng2)+math.sin(lat1)*math.sin(lat2))
def calculate_Spherical_distance_L_to_L(L_start, L_end):
LL = []
for i in range(len(L_start)):
L = []
for j in range(len(L_end)):
L.append(calculate_Spherical_distance(L_start[i][1], L_start[i][2], L_end[j][1], L_end[j][2]))
LL.append(copy.deepcopy(L))
return LL
def L_vehicles_generate(warehouses, vehicles):
L_vehicles = []
for ware in warehouses:
v_per = ware[3]*ware[4]
vs = copy.deepcopy(vehicles)
if v_per != 0:
vs[0][4] = int(vs[0][4]/v_per)
vs[0][5] = int(vs[0][5]/v_per)
vs[1][4] = int(vs[1][4]/v_per)
vs[1][5] = int(vs[1][5]/v_per)
L_vehicles.append(vs)
return L_vehicles
def solving_violence(X_set, Y_set, w11, w12, w21, w22):
x_best = [X_set[i][0] for i in range(11)]
y_best = [Y_set[i][0] for i in range(11)]
c_best = sum([(w11+w12)*x_best[i] for i in range(11)]) + sum([(w21+w22)*y_best[i] for i in range(11)])
"""复杂度上限，超过上限时不计算"""
limit_iteration = 1e6
num_iteration = 1
for i in range(len(X_set)):
num_iteration *= len(X_set[i])
if num_iteration > limit_iteration:
return [x_best, y_best]
for i0 in range(len(X_set[0])):
x0 = X_set[0][i0]
y0 = Y_set[0][i0]
for i1 in range(len(X_set[1])):
x1 = X_set[1][i1]
y1 = Y_set[1][i1]
for i2 in range(len(X_set[2])):
x2 = X_set[2][i2]
y2 = Y_set[2][i2]
for i3 in range(len(X_set[3])):
x3 = X_set[3][i3]
y3 = Y_set[3][i3]
for i4 in range(len(X_set[4])):
x4 = X_set[4][i4]
y4 = Y_set[4][i4]
for i5 in range(len(X_set[5])):
x5 = X_set[5][i5]
y5 = Y_set[5][i5]
for i6 in range(len(X_set[6])):
x6 = X_set[6][i6]
y6 = Y_set[6][i6]
for i7 in range(len(X_set[7])):
x7 = X_set[7][i7]
y7 = Y_set[7][i7]
for i8 in range(len(X_set[8])):
x8 = X_set[8][i8]
y8 = Y_set[8][i8]
for i9 in range(len(X_set[9])):
x9 = X_set[9][i9]
y9 = Y_set[9][i9]
for i10 in range(len(X_set[10])):
x10 = X_set[10][i10]
y10 = Y_set[10][i10]
c = (w11+w12)*(x0+x1+x2+x3+x4+x5+x6+x7+x8+x9+x10) + (w21+w22)*(y0+y1+y2+y3+y4+y5+y6+y7+y8+y9+y10)
if c_best > c:
c_best = c
x_best = [x0,x1,x2,x3,x4,x5,x6,x7,x8,x9,x10]
y_best = [y0,y1,y2,y3,y4,y5,y6,y7,y8,y9,y10]
return [x_best, y_best]
def sub_calculate_cost(nums, dis, vehicles, L_frequency, is_tray):
if dis <= L_frequency[1][0]:
frequency = L_frequency[0][1]
elif dis <= L_frequency[2][0]:
frequency = L_frequency[1][1]
else:
frequency = L_frequency[2][1]
w11 = (vehicles[0][1] + vehicles[0][2])/frequency
w12 = vehicles[0][3]*dis*2
w21 = (vehicles[1][1] + vehicles[1][2])/frequency
w22 = vehicles[1][3]*dis*2
if is_tray:
A = [vehicles[0][5], vehicles[1][5]]
else:
A = [vehicles[0][4], vehicles[1][4]]
X_set = []
Y_set = []
for i in range(11):
Xi_set = []
Yi_set = []
max_xi = math.ceil(nums[i]/A[0])
max_yi = math.ceil(nums[i]/A[1])
if max_xi <= max_yi:
for j in range(max_xi+1):
Xi_set.append(j)
Yi_set.append(max(math.ceil((nums[i]-j*A[0])/A[1]),0))
else:
for j in range(max_yi+1):
Yi_set.append(j)
Xi_set.append(max(math.ceil((nums[i]-j*A[1])/A[0]),0))
X_set.append(Xi_set)
Y_set.append(Yi_set)
result_violence = solving_violence(X_set, Y_set, w11, w12, w21, w22)
x_best = result_violence[0]
y_best = result_violence[1]
cost_fixed = w11*sum(x_best) + w21*sum(y_best)
cost_variable = w12*sum(x_best) + w22*sum(y_best)
cost_all = cost_fixed + cost_variable
return [cost_all, cost_variable, cost_fixed] + x_best + y_best
def output_original(cost_all, cost_ware, ware_to_first, name):
if os.path.exists(name):
os.remove(name)
f1 = open(name, "a" , encoding = 'gb18030')
f1.write("总成本\n")
f1.write(str(cost_all) + "\n")
f1.write("\n")
f1.write("仓成本\n")
f1.write("仓,仓总单量,仓总成本,")
f1.write("单10_11,单1,单2,单3,单4,单5,单6,单7,单8,单9,单10,")
f1.write("cost_10_11,cost_1,cost_2,cost_3,cost_4,cost_5,cost_6,cost_7,cost_8,cost_9,cost_10\n")
for i in range(len(cost_ware)):
for j in range(len(cost_ware[i])):
f1.write(str(cost_ware[i][j]))
f1.write(',')
f1.write('\n')
f1.write("\n")
f1.write("仓至首分拣成本\n")
f1.write("接货仓,首分拣,距离,")
f1.write("总运单量,单10_11,单1,单2,单3,单4,单5,单6,单7,单8,单9,单10,")
f1.write("运输成本,固定成本,变动成本,")
f1.write("9m6_10_11,9m6_1,9m6_2,9m6_3,9m6_4,9m6_5,9m6_6,9m6_7,9m6_8,9m6_9,9m6_10,")
f1.write("7m6_10_11,7m6_1,7m6_2,7m6_3,7m6_4,7m6_5,7m6_6,7m6_7,7m6_8,7m6_9,7m6_10\n")
for i in range(len(ware_to_first)):
for j in range(len(ware_to_first[i])):
f1.write(str(ware_to_first[i][j]))
f1.write(',')
f1.write('\n')
f1.close()
def txt_process(s):
if os.path.exists("process.txt"):
os.remove("process.txt")
f1 = open("process.txt", "a" , encoding = 'gb18030')
f1.write(s)
f1.close()
def calculate_cost_original(data, data_base, parameter, output_path):
warehouses = data_base[0]
delvs_first = data_base[1]
D_ware = {}
D_delv = {}
for i in range(len(warehouses)):
D_ware[warehouses[i][0]] = i
for i in range(len(delvs_first)):
D_delv[delvs_first[i][0]] = i
"""初始化单量矩阵"""
order = []
for i in range(11):
order.append(matrix_zero(len(warehouses),len(delvs_first)))
for d in data:
for i in range(11):
order[i][D_ware[d[0]]][D_delv[d[3]]] += d[10+i]
Dis = calculate_Spherical_distance_L_to_L(warehouses, delvs_first)
"""单量、距离归纳好了，开始计算原始成本"""
cost_ware = []
for i in range(len(warehouses)):
nums_ware_i = []
for k in range(len(order)):
nums_k = 0
for j in range(len(delvs_first)):
nums_k += order[k][i][j]
nums_ware_i.append(nums_k)
cost_ware_i = [nums_ware_i[k]*parameter[2]/parameter[0] for k in range(len(nums_ware_i))]
cost_ware.append([warehouses[i][0], sum(nums_ware_i), sum(cost_ware_i)] + nums_ware_i + cost_ware_i)
ware_to_first = []
L_vehicles = L_vehicles_generate(warehouses, parameter[7])
flag = 0
for i in range(len(warehouses)):
for j in range(len(delvs_first)):
nums = []
for k in range(len(order)):
nums.append(order[k][i][j])
"""根据单量、距离、车辆、频率计算成本"""
if sum(nums) == 0 or Dis[i][j] == 0:
ware_to_first.append([warehouses[i][0], delvs_first[j][0], Dis[i][j], sum(nums)] + nums + [0 for i in range(25)])
else:
flag += 1
print("原始成本第"+str(flag)+"条路线成本测算")
#txt_process("原始成本第"+str(flag)+"条路线成本测算")
cost_and_vehicle = sub_calculate_cost(nums, Dis[i][j], L_vehicles[i], parameter[9], True)
ware_to_first.append([warehouses[i][0], delvs_first[j][0], Dis[i][j], sum(nums)] + nums + cost_and_vehicle)
cost_all = 0
for i in range(len(warehouses)):
cost_all += cost_ware[i][2]
for i in range(len(ware_to_first)):
cost_all += ware_to_first[i][15]
output_original(cost_all, cost_ware, ware_to_first, output_path+"\\original.csv")
return
"""优化计算部分"""
"""
def clusters_plot(warehouses, L_w):
x = []
y = []
col = []
for i in range(len(warehouses)):
x.append(warehouses[i][1])
y.append(warehouses[i][2])
col.append(L_w[i])
plt.scatter(x,y,c = col)
"""
def solving_violence_10_11(Xi_set, Yi_set, w11, w12, w21, w22):
x_best = Xi_set[0]
y_best = Yi_set[0]
c_best = (w11+w12)*x_best + (w21+w22)*y_best
for i in range(1,len(Xi_set)):
x = Xi_set[i]
y = Yi_set[i]
c = (w11+w12)*x + (w21+w22)*y
if c_best > c:
c_best = c
x_best = x
y_best = y
return [x_best, y_best]
def sub_calculate_cost_10_11(num, dis, vehicles, L_frequency, is_tray):
#sub_calculate_cost_10_11(order_ware_to_first[i][j], Dis_ware_to_first[i][j], L_vehicles[i], parameter[9], True)
if dis <= L_frequency[1][0]:
frequency = L_frequency[0][1]
elif dis <= L_frequency[2][0]:
frequency = L_frequency[1][1]
else:
frequency = L_frequency[2][1]
w11 = (vehicles[0][1] + vehicles[0][2])/frequency
w12 = vehicles[0][3]*dis*2
w21 = (vehicles[1][1] + vehicles[1][2])/frequency
w22 = vehicles[1][3]*dis*2
if is_tray:
A = [vehicles[0][5], vehicles[1][5]]
else:
A = [vehicles[0][4], vehicles[1][4]]
Xi_set = []
Yi_set = []
max_xi = math.ceil(num/A[0])
max_yi = math.ceil(num/A[1])
if max_xi <= max_yi:
for j in range(max_xi+1):
Xi_set.append(j)
Yi_set.append(max(math.ceil((num-j*A[0])/A[1]),0))
else:
for j in range(max_yi+1):
Yi_set.append(j)
Xi_set.append(max(math.ceil((num-j*A[1])/A[0]),0))
result_violence = solving_violence_10_11(Xi_set, Yi_set, w11, w12, w21, w22)
x_best = result_violence[0]
y_best = result_violence[1]
cost_fixed = w11*x_best + w21*y_best
cost_variable = w12*x_best + w22*y_best
cost_all = cost_fixed + cost_variable
return [cost_all, cost_variable, cost_fixed, x_best, y_best]
def solving_violence_ware_to_front(X_set, Y_set, w11, w12, w21, w22):
x_best = [X_set[i][0] for i in range(10)]
y_best = [Y_set[i][0] for i in range(10)]
c_best = sum([(w11+w12)*x_best[i] for i in range(10)]) + sum([(w21+w22)*y_best[i] for i in range(10)])
"""复杂度上限，超过上限时不计算"""
limit_iteration = 1e6
num_iteration = 1
for i in range(len(X_set)):
num_iteration *= len(X_set[i])
if num_iteration > limit_iteration:
return [x_best, y_best]
for i0 in range(len(X_set[0])):
x0 = X_set[0][i0]
y0 = Y_set[0][i0]
for i1 in range(len(X_set[1])):
x1 = X_set[1][i1]
y1 = Y_set[1][i1]
for i2 in range(len(X_set[2])):
x2 = X_set[2][i2]
y2 = Y_set[2][i2]
for i3 in range(len(X_set[3])):
x3 = X_set[3][i3]
y3 = Y_set[3][i3]
for i4 in range(len(X_set[4])):
x4 = X_set[4][i4]
y4 = Y_set[4][i4]
for i5 in range(len(X_set[5])):
x5 = X_set[5][i5]
y5 = Y_set[5][i5]
for i6 in range(len(X_set[6])):
x6 = X_set[6][i6]
y6 = Y_set[6][i6]
for i7 in range(len(X_set[7])):
x7 = X_set[7][i7]
y7 = Y_set[7][i7]
for i8 in range(len(X_set[8])):
x8 = X_set[8][i8]
y8 = Y_set[8][i8]
for i9 in range(len(X_set[9])):
x9 = X_set[9][i9]
y9 = Y_set[9][i9]
c = (w11+w12)*(x0+x1+x2+x3+x4+x5+x6+x7+x8+x9) + (w21+w22)*(y0+y1+y2+y3+y4+y5+y6+y7+y8+y9)
if c_best > c:
c_best = c
x_best = [x0,x1,x2,x3,x4,x5,x6,x7,x8,x9]
y_best = [y0,y1,y2,y3,y4,y5,y6,y7,y8,y9]
return [x_best, y_best]
def sub_calculate_cost_ware_to_front(nums, dis, vehicles, L_frequency, is_tray):
if dis <= L_frequency[1][0]:
frequency = L_frequency[0][1]
elif dis <= L_frequency[2][0]:
frequency = L_frequency[1][1]
else:
frequency = L_frequency[2][1]
w11 = (vehicles[0][1] + vehicles[0][2])/frequency
w12 = vehicles[0][3]*dis*2
w21 = (vehicles[1][1] + vehicles[1][2])/frequency
w22 = vehicles[1][3]*dis*2
if is_tray:
A = [vehicles[0][5], vehicles[1][5]]
else:
A = [vehicles[0][4], vehicles[1][4]]
X_set = []
Y_set = []
for i in range(10):
Xi_set = []
Yi_set = []
max_xi = math.ceil(nums[i]/A[0])
max_yi = math.ceil(nums[i]/A[1])
if max_xi <= max_yi:
for j in range(max_xi+1):
Xi_set.append(j)
Yi_set.append(max(math.ceil((nums[i]-j*A[0])/A[1]),0))
else:
for j in range(max_yi+1):
Yi_set.append(j)
Xi_set.append(max(math.ceil((nums[i]-j*A[1])/A[0]),0))
X_set.append(Xi_set)
Y_set.append(Yi_set)
result_violence = solving_violence_ware_to_front(X_set, Y_set, w11, w12, w21, w22)
x_best = result_violence[0]
y_best = result_violence[1]
cost_fixed = w11*sum(x_best) + w21*sum(y_best)
cost_variable = w12*sum(x_best) + w22*sum(y_best)
cost_all = cost_fixed + cost_variable
return [cost_all, cost_variable, cost_fixed] + x_best + y_best
def vehicles_front_to_second_generate(L_vehicles):
vehicles_front_to_second = copy.deepcopy(L_vehicles[0])
vehicles_front_to_second[0][4] = sum([L_vehicles[i][0][4] for i in range(len(L_vehicles))])/len(L_vehicles)
vehicles_front_to_second[0][5] = sum([L_vehicles[i][0][5] for i in range(len(L_vehicles))])/len(L_vehicles)
vehicles_front_to_second[1][4] = sum([L_vehicles[i][1][4] for i in range(len(L_vehicles))])/len(L_vehicles)
vehicles_front_to_second[1][5] = sum([L_vehicles[i][1][5] for i in range(len(L_vehicles))])/len(L_vehicles)
return vehicles_front_to_second
def order_distribute(order_front_to_second, vehicles_front_to_second, idex_second):
num_big = vehicles_front_to_second[0][4]
num_small = vehicles_front_to_second[1][4]
order_direct = []
order_left = []
rate = []
for i in range(len(order_front_to_second)):
order_direct.append(matrix_zero(len(order_front_to_second[i]),len(order_front_to_second[i][0])))
order_left.append(matrix_zero(len(order_front_to_second[i]),len(order_front_to_second[i][0])))
rate.append(matrix_zero(len(order_front_to_second[i]),len(order_front_to_second[i][0])))
for i in range(len(order_front_to_second)):
for j in range(len(order_front_to_second[i])):
for k in range(len(order_front_to_second[i][j])):
left = order_front_to_second[i][j][k]%num_big%num_small
if k == idex_second:
left = order_front_to_second[i][j][k]
order_direct[i][j][k] = order_front_to_second[i][j][k] - left
order_left[i][j][k] = left
if order_direct[i][j][k] > 0:
rate[i][j][k] = left/order_direct[i][j][k]
return [order_direct, order_left, rate]
def output_front(cost_all, num_ware_10_11, cost_ware_10_11, num_front, cost_front, result_ware_to_first,
result_ware_to_front, result_front_to_second, result_front_to_first, warehouses, delvs_first,
delvs_second, delvs_front, name):
if os.path.exists(name):
os.remove(name)
f1 = open(name, "a" , encoding = 'gb18030')
f1.write("成本项,成本,单量\n")
f1.write("总成本," + str(cost_all) + "\n")
f1.write("仓," + str(sum(cost_ware_10_11)) + "," + str(sum(num_ware_10_11)) + "\n")
f1.write("前置分拣," + str(sum(cost_front)) + "," + str(sum(num_front)) + "\n")
f1.write("仓-首分拣," + str(sum([result_ware_to_first[i][4] for i in range(len(result_ware_to_first))])) + "," +
str(sum([result_ware_to_first[i][3] for i in range(len(result_ware_to_first))])) + "\n")
f1.write("仓-前置分拣," + str(sum([result_ware_to_front[i][14] for i in range(len(result_ware_to_front))])) + "," +
str(sum([result_ware_to_front[i][3] for i in range(len(result_ware_to_front))])) + "\n")
f1.write("前置分拣-首分拣," + str(sum([result_front_to_first[i][14] for i in range(len(result_front_to_first))])) + "," +
str(sum([result_front_to_first[i][3] for i in range(len(result_front_to_first))])) + "\n")
f1.write("前置分拣-二分拣," + str(-sum([result_front_to_second[i][13] for i in range(len(result_front_to_second))])) + "," +
str(sum([result_front_to_second[i][2] for i in range(len(result_front_to_second))])) + "\n")
f1.write("\n")
f1.write("仓成本\n")
f1.write("仓,仓总成本,仓总单量\n")
for i in range(len(warehouses)):
f1.write(warehouses[i][0]+","+str(cost_ware_10_11[i])+","+str(num_ware_10_11[i])+"\n")
f1.write("\n")
f1.write("前置分拣成本\n")
f1.write("前置分拣,前置分拣总成本,前置分拣总单量\n")
for i in range(len(delvs_front)):
f1.write(delvs_front[i][0]+","+str(cost_front[i])+","+str(num_front[i])+"\n")
f1.write("\n")
f1.write("仓-首分拣\n")
f1.write("接货仓,首分拣,距离,总运单量,运输成本,固定成本,变动成本,9m6_10_11,7m6_10_11\n")
for i in range(len(result_ware_to_first)):
for j in range(len(result_ware_to_first[i])):
f1.write(str(result_ware_to_first[i][j]))
f1.write(',')
f1.write('\n')
f1.write("\n")
f1.write("仓-前置分拣\n")
f1.write("接货仓,前置分拣,距离,")
f1.write("总运单量,单1,单2,单3,单4,单5,单6,单7,单8,单9,单10,")
f1.write("运输成本,固定成本,变动成本,")
f1.write("9m6_1,9m6_2,9m6_3,9m6_4,9m6_5,9m6_6,9m6_7,9m6_8,9m6_9,9m6_10,")
f1.write("7m6_1,7m6_2,7m6_3,7m6_4,7m6_5,7m6_6,7m6_7,7m6_8,7m6_9,7m6_10\n")
for i in range(len(result_ware_to_front)):
for j in range(len(result_ware_to_front[i])):
f1.write(str(result_ware_to_front[i][j]))
f1.write(',')
f1.write('\n')
f1.write("\n")
f1.write("前置分拣-首分拣\n")
f1.write("前置分拣,首分拣,距离,")
f1.write("总运单量,单1,单2,单3,单4,单5,单6,单7,单8,单9,单10,")
f1.write("运输成本,固定成本,变动成本,")
f1.write("9m6_1,9m6_2,9m6_3,9m6_4,9m6_5,9m6_6,9m6_7,9m6_8,9m6_9,9m6_10,")
f1.write("7m6_1,7m6_2,7m6_3,7m6_4,7m6_5,7m6_6,7m6_7,7m6_8,7m6_9,7m6_10\n")
for i in range(len(result_front_to_first)):
for j in range(len(result_front_to_first[i])):
f1.write(str(result_front_to_first[i][j]))
f1.write(',')
f1.write('\n')
f1.write("\n")
f1.write("前置分拣-二分拣\n")
f1.write("前置分拣,二分拣,")
f1.write("总运单量,单1,单2,单3,单4,单5,单6,单7,单8,单9,单10,")
f1.write("降低成本,1,2,3,4,5,6,7,8,9,10\n")
for i in range(len(result_front_to_second)):
for j in range(len(result_front_to_second[i])):
f1.write(str(result_front_to_second[i][j]))
f1.write(',')
f1.write('\n')
f1.close()
def sub_calculate_cost_front(data, warehouses, delvs_first, delvs_second, parameter, delvs_front, flag_print, output_path):
#根据仓库数据调整每种车型的规格
L_vehicles = L_vehicles_generate(warehouses, parameter[7])
#创建下标字典
D_ware = {}
D_first = {}
D_front = {}
D_second = {}
for i in range(len(warehouses)):
D_ware[warehouses[i][0]] = i
for i in range(len(delvs_first)):
D_first[delvs_first[i][0]] = i
for i in range(len(delvs_front)):
D_front[delvs_front[i][0]] = i
for i in range(len(delvs_second)):
D_second[delvs_second[i][0]] = i
#根据距离就近选择前置分拣
L_w = []
for i in range(len(warehouses)):
idex = 0
for j in range(len(delvs_front)):
if calculate_Spherical_distance(warehouses[i][1], warehouses[i][2], delvs_front[idex][1], delvs_front[idex][2]) > calculate_Spherical_distance(warehouses[i][1], warehouses[i][2], delvs_front[j][1], delvs_front[j][2]):
idex = j
L_w.append(idex)
#打印聚类结果
#clusters_plot(warehouses, L_w)
#前置分拣添加至运单数据
data1 = copy.deepcopy(data)
for i in range(len(data1)):
data1[i] = data1[i] + delvs_front[L_w[D_ware[data1[i][0]]]]
#统计10-11点的仓发出单量及结果仓成本
num_ware_10_11 = [0 for i in range(len(warehouses))]
for d in data1:
num_ware_10_11[D_ware[d[0]]] += d[10]
cost_ware_10_11 = [num_ware_10_11[i]*parameter[2]/parameter[1] for i in range(len(num_ware_10_11))]
#前置分拣成本
num_front = [0 for i in range(len(delvs_front))]
for d in data1:
num_front[D_front[d[21]]] += sum(d[11:21])
cost_front = [num_front[i]*parameter[5] for i in range(len(num_front))]
#仓到首分拣成本
order_ware_to_first = matrix_zero(len(warehouses),len(delvs_first))
for d in data1:
order_ware_to_first[D_ware[d[0]]][D_first[d[3]]] += d[10]
Dis_ware_to_first = calculate_Spherical_distance_L_to_L(warehouses, delvs_first)
result_ware_to_first = []
for i in range(len(warehouses)):
for j in range(len(delvs_first)):
if order_ware_to_first[i][j] == 0 or Dis_ware_to_first[i][j] == 0:
"""仓、首分拣、距离、10_11单量、运输成本、固定成本、变动成本、9m6_10_11、7m6_10_11"""
result_ware_to_first.append([warehouses[i][0], delvs_first[j][0], Dis_ware_to_first[i][j], order_ware_to_first[i][j]] + [0 for i in range(5)])
else:
cost_and_vehicle = sub_calculate_cost_10_11(order_ware_to_first[i][j], Dis_ware_to_first[i][j], L_vehicles[i], parameter[9], True)
result_ware_to_first.append([warehouses[i][0], delvs_first[j][0], Dis_ware_to_first[i][j], order_ware_to_first[i][j]] + cost_and_vehicle)
#仓到前置分拣成本
order_ware_to_front = []
for i in range(10):
order_ware_to_front.append(matrix_zero(len(warehouses),len(delvs_front)))
for d in data1:
for i in range(10):
order_ware_to_front[i][D_ware[d[0]]][D_front[d[21]]] += d[11+i]
Dis_ware_to_front = calculate_Spherical_distance_L_to_L(warehouses, delvs_front)
result_ware_to_front = []
flag = 0
for i in range(len(warehouses)):
for j in range(len(delvs_front)):
nums = []
for k in range(len(order_ware_to_front)):
nums.append(order_ware_to_front[k][i][j])
"""根据单量、距离、车辆、频率计算成本"""
if sum(nums) == 0 or Dis_ware_to_front[i][j] == 0:
result_ware_to_front.append([warehouses[i][0], delvs_front[j][0], Dis_ware_to_front[i][j], sum(nums)] + nums + [0 for i in range(23)])
else:
flag += 1
print("仓到前置分拣优化成本第"+str(flag)+"条路线成本测算")
#txt_process("仓到前置分拣优化成本第"+str(flag)+"条路线成本测算")
cost_and_vehicle = sub_calculate_cost_ware_to_front(nums, Dis_ware_to_front[i][j], L_vehicles[i], parameter[9], False)
result_ware_to_front.append([warehouses[i][0], delvs_front[j][0], Dis_ware_to_front[i][j], sum(nums)] + nums + cost_and_vehicle)
#前置分拣至二级分拣减少成本及明细
order_front_to_second = []
for i in range(10):
order_front_to_second.append(matrix_zero(len(delvs_front),len(delvs_second)))
for d in data1:
for i in range(10):
order_front_to_second[i][D_front[d[21]]][D_second[d[6]]] += d[11+i]
vehicles_front_to_second = vehicles_front_to_second_generate(L_vehicles)
result_distribute = order_distribute(order_front_to_second, vehicles_front_to_second, D_second[""])
order_direct = result_distribute[0]
result_front_to_second = []
for i in range(len(delvs_front)):
for j in range(len(delvs_second)):
nums = []
costs = []
for k in range(len(order_direct)):
nums.append(order_direct[k][i][j])
costs.append(order_direct[k][i][j]*parameter[6])
result_front_to_second.append([delvs_front[i][0], delvs_second[j][0], sum(nums)] + nums +
[sum(costs)] + costs)
#前置分拣至首分拣成本
rate_left = result_distribute[2]
data2 = copy.deepcopy(data1)
for i in range(len(data2)):
for j in range(10):
data2[i][11+j] = data2[i][11+j]*rate_left[j][D_front[data2[i][21]]][D_second[data2[i][6]]]
order_front_to_first = []
for i in range(10):
order_front_to_first.append(matrix_zero(len(delvs_front),len(delvs_first)))
for d in data2:
for i in range(10):
order_front_to_first[i][D_front[d[21]]][D_first[d[3]]] += d[11+i]
Dis_front_to_first = calculate_Spherical_distance_L_to_L(delvs_front, delvs_first)
result_front_to_first = []
flag = 0
for i in range(len(delvs_front)):
for j in range(len(delvs_first)):
nums = []
for k in range(len(order_front_to_first)):
nums.append(order_front_to_first[k][i][j])
"""根据单量、距离、车辆、频率计算成本"""
if sum(nums) == 0 or Dis_front_to_first[i][j] == 0:
result_front_to_first.append([delvs_front[i][0], delvs_first[j][0], Dis_front_to_first[i][j], sum(nums)] + nums + [0 for i in range(23)])
else:
flag += 1
print("前置分拣到首分拣优化成本第"+str(flag)+"条路线成本测算")
#txt_process("前置分拣到首分拣优化成本第"+str(flag)+"条路线成本测算")
cost_and_vehicle = sub_calculate_cost_ware_to_front(nums, Dis_front_to_first[i][j], vehicles_front_to_second, parameter[9], True)
result_front_to_first.append([delvs_front[i][0], delvs_first[j][0], Dis_front_to_first[i][j], sum(nums)] + nums + cost_and_vehicle)
cost_all = 0
cost_all += sum(cost_ware_10_11)
cost_all += sum(cost_front)
for i in range(len(result_ware_to_first)):
cost_all += result_ware_to_first[i][4]
for i in range(len(result_ware_to_front)):
cost_all += result_ware_to_front[i][14]
for i in range(len(result_front_to_first)):
cost_all += result_front_to_first[i][14]
for i in range(len(result_front_to_second)):
cost_all -= result_front_to_second[i][13]
if flag_print:
output_front(cost_all, num_ware_10_11, cost_ware_10_11, num_front, cost_front,
result_ware_to_first, result_ware_to_front, result_front_to_second,
result_front_to_first, warehouses, delvs_first, delvs_second,
delvs_front, output_path + "\\front.csv")
return cost_all
def delvs_possible_generate(warehouses, density):
lat_min = warehouses[0][1]
lat_max = warehouses[0][1]
lng_min = warehouses[0][2]
lng_max = warehouses[0][2]
for ware in warehouses:
if lat_min > ware[1]:
lat_min = ware[1]
if lat_max < ware[1]:
lat_max = ware[1]
if lng_min > ware[2]:
lng_min = ware[2]
if lng_max < ware[2]:
lng_max = ware[2]
delvs = []
for i in range(density+1):
for j in range(density+1):
delvs.append(["delv_"+str(i*(density+1)+j),
lat_min * (density - i) / density + lat_max * i / density,
lng_min * (density - j) / density + lng_max * j / density])
return delvs
def order_count(warehouses, data):
D_ware = {}
for i in range(len(warehouses)):
D_ware[warehouses[i][0]] = i
order_sum = [0 for i in range(len(warehouses))]
for d in data:
order_sum[D_ware[d[0]]] += sum(d[11:])
return order_sum
def distance_class(class1, class2):
c1 = [0,0]
c2 = [0,0]
for i in range(len(class1)):
c1[0] += class1[i][1]
c1[1] += class1[i][2]
c1[0] /= len(class1)
c1[1] /= len(class1)
for i in range(len(class2)):
c2[0] += class2[i][1]
c2[1] += class2[i][2]
c2[0] /= len(class2)
c2[1] /= len(class2)
return calculate_Spherical_distance(c1[0],c1[1],c2[0],c2[1])
def clustering(warehouses, order_sum_warehouse, limit):
clusters = []
for i in range(len(warehouses)):
clusters.append([copy.deepcopy(warehouses[i][:3])])
L_w = []
for i in range(len(warehouses)):
L_w.append([order_sum_warehouse[i]])
while True:
print("聚类中，目前size="+str(len(clusters))+", "+datetime.datetime.now().strftime("%Y-%m-%d %H:%M:%S"))
#txt_process("聚类中，目前size="+str(len(clusters))+", "+datetime.datetime.now().strftime("%Y-%m-%d %H:%M:%S"))
idex1 = 0
idex2 = 1
flag = 0
for i in range(len(clusters)-1):
for j in range(i+1, len(clusters)):
if (distance_class(clusters[i],clusters[j]) < distance_class(clusters[idex1],clusters[idex2]) and
sum(L_w[i]) + sum(L_w[j]) <= limit):
idex1 = i
idex2 = j
flag = 1
if flag == 0 and sum(L_w[idex1]) + sum(L_w[idex2]) > limit:
break
clusters_new = []
L_w_new = []
for i in range(len(clusters)):
if i == idex1 or i == idex2:
continue
clusters_new.append(copy.deepcopy(clusters[i]))
L_w_new.append(copy.deepcopy(L_w[i]))
clusters_new.append(copy.deepcopy(clusters[idex1]+clusters[idex2]))
L_w_new.append(copy.deepcopy(L_w[idex1]+L_w[idex2]))
clusters = clusters_new
L_w = L_w_new
delvs_front = []
for i in range(len(clusters)):
idex = 0
for j in range(len(clusters[i])):
if L_w[i][idex] < L_w[i][j]:
idex = j
delvs_front.append(["delv_"+str(i)] + clusters[i][idex][1:])
return delvs_front
def calculate_cost_front(data, data_base, parameter, output_path):
warehouses = data_base[0]
delvs_first = data_base[1]
delvs_second = data_base[2]
#delvs_possible = delvs_possible_generate(warehouses, 20)
limit = parameter[3]*parameter[4]
order_sum_warehouse = order_count(warehouses, data)
delvs_front = clustering(warehouses, order_sum_warehouse, limit)
"""
delvs_front = [["北京周边",39.69139,116.5659],
["济南",36.82950057,117.0835159],
["青岛",36.11402742,120.103142],
["石家庄太原临汾",37.4018790899999,114.063889899999],
["呼和浩特",40.74075379,111.90701415]]
"""
cost_front = sub_calculate_cost_front(data, warehouses, delvs_first, delvs_second, parameter, delvs_front, True, output_path)
"""
result_greedy = calculate_greedy(warehouses, delvs_first, delvs_second, parameter, data, 10)
delvs_best_local = result_greedy[0]
front_map_best_local = result_greedy[1]
"""
return
def validCaptcha(code):
"""
验证License是否正确
"""
if code is None or code.strip() == '' :
return False
temp = code.split(':')
erp = temp[0];
captach = temp[1];
w=wmi.WMI()
cpu_id=w.Win32_Processor()[0].ProcessorId
licenseCode = erp + '::' + cpu_id
m = hashlib.md5()
m.update(licenseCode.encode("utf-8"))
md5 = m.hexdigest().upper()
return captach == md5
if __name__ == "__main__":
parameter_path = r'配置文件.csv'
data_path = r"数据.csv"
output_path = os.getcwd()
password = ""
print(sys.argv)
print([parameter_path,data_path,output_path,password])
if len(sys.argv) >= 2:
parameter_path = sys.argv[1]
if len(sys.argv) >= 3:
data_path = sys.argv[2]
if len(sys.argv) >= 4:
output_path = sys.argv[3]
if len(sys.argv) >= 5:
password = sys.argv[4]
#password = 'dbwcyang:40A7DC7B71AE12684B97D9DAE2DBA6F4'
print('License['+password+']验证...')
#验证license
if not validCaptcha(password):
print('License验证失败，无权使用该程序')
sys.exit(0)
print('License['+password+']验证成功')
print([parameter_path,data_path,output_path,password])
print("读取参数")
"""
原人效(单/人/天)、新人效(只考虑10-11点)、人均工资（元/天/人）、坪效(单/平方米)、
最大面积(平方米)、操作成本(元/单)、直发二级分拣节省的操作成本(元/单)、车辆配置、
依据单量百分比确定个数、车辆复用规则
"""
parameter = read_parameter_csv(open(parameter_path))
print("读取数据")
data = read_data_csv(open(data_path))
#根据parameter中接货仓的信息筛选订单
data = data_clean(data, parameter)
#仓、首分拣、二分拣
data_base = data_base_read(data, parameter)
print("开始计算")
#计算原始成本
calculate_cost_original(data, data_base, parameter, output_path)
#计算优化成本
calculate_cost_front(data, data_base, parameter, output_path)
#txt_process("end")
print("end")
2万+
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
