Add files via upload

Author: dungtran209

code/1. GG API.py

@@ -0,0 +1,47 @@
+#!/usr/bin/env python
+# coding: utf-8
+
+import os
+import pandas as pd
+import time
+
+s = time.time()
+# Importing googlemaps module
+import googlemaps
+
+# Requires API key
+gmaps = googlemaps.Client(key='AIzaSyBSzEO6ecgV6Kr8Vg-5dmb2ww4YFhyOM6s')
+
+dir_name = os.path.dirname(os.path.realpath('__file__'))
+file ="9Aug19"
+
+file_name = os.path.join(dir_name, 'Real Dataset\\',file +'.csv')
+df = pd.read_csv(file_name, encoding ='latin1', error_bad_lines=False)
+
+d = {'Name': 'first', 'Post Code': 'first', 'Date': 'first', 'Reference': 'first', 'Type': 'first', 'Value' : 'sum'}
+df_new = df.groupby('No', as_index=False).aggregate(d).reindex(columns=df.columns)
+print(df_new)
+
+cust_size = df_new.shape[0]
+
+postcode = df_new['Post Code']
+
+dist_matrix = pd.DataFrame(0, index=range(cust_size), columns=range(cust_size))
+time_matrix = pd.DataFrame(0, index=range(cust_size), columns=range(cust_size))
+
+for i in range(cust_size):
+    for j in range(cust_size):
+        if i!= j:
+            my_dist = gmaps.distance_matrix(postcode[i],postcode[j])['rows'][0]['elements'][0]
+            #print(i, j, my_dist)
+            dist_matrix[i][j] = my_dist['distance']['value']
+            time_matrix[i][j] = my_dist['duration']['value']
+
+e =time.time()
+
+print(e-s)
+dist_matrix.to_csv('Real Dataset\\9AugDist.csv', index=False, header=False)
+time_matrix.to_csv('Real Dataset\\9AugTime.csv', index=False, header=False)
+
+
+

code/2. Test_ExactMethod.py

@@ -0,0 +1,155 @@
+#!/usr/bin/env python
+# coding: utf-8
+
+import os
+import numpy as np
+import pandas as pd
+import sklearn as sk
+import matplotlib.pyplot as plt
+import time
+
+from scipy.spatial import distance_matrix
+
+# In[14]:
+
+
+import sys
+import docplex.mp
+
+# try:
+#     import docplex.mp
+# except:
+#     if hasattr(sys, 'real_prefix'):
+#         get_ipython().system('pip install docplex -q')
+#     else:
+#         get_ipython().system('pip install --user docplex -q')
+
+from docplex.mp.model import Model
+
+dir_name = os.path.dirname(os.path.realpath('__file__'))
+
+# folder = ["R101", "R102", "R103", "R104", "R105", "R106", "R107", "R108", "R109", "R110", "R111", "R112",
+#           "R201", "R202", "R203", "R204", "R205", "R206", "R207", "R208", "R209", "R210", "R211",
+#           "RC101", "RC102", "RC103", "RC104", "RC105", "RC106", "RC107", "RC108",
+#           "RC201", "RC202", "RC203", "RC204", "RC205", "RC206", "RC207", "RC208",
+#           "C201", "C202", "C203", "C204", "C205", "C206", "C207", "C208",
+#           "C101", "C102", "C103", "C104", "C105", "C106", "C107", "C108", "C109"]
+
+file_names = ["R101", "R106", "C101", "C106", "RC101", "RC106", "R201", "R206", "C201", "C206", "RC201", "RC206"]
+
+K = 10000
+
+for file in file_names:
+    file_name = os.path.join(dir_name, 'Sample Dataset\\', file + '.csv')
+    df = pd.read_csv(file_name, encoding='latin1', error_bad_lines=False);
+
+    df = df[0:26]
+
+    cust_size = df.shape[0] - 1
+    # print('Number of customers:', cust_size)
+    df.head()
+
+    n = cust_size
+    Q = df['CAPACITY'][0]
+    C = [i for i in range(1, n + 1)]
+    Cc = [0] + C + [n + 1]
+    V = [i for i in range(1, 26)]
+
+    df2 = df.iloc[:, 1:3]
+    df2.loc[n + 1, :] = df2.loc[0, :]
+
+    dist_matrix = pd.DataFrame(distance_matrix(df2.values, df2.values), index=df2.index, columns=df2.index)
+
+    # dist_matrix.head()
+
+    time_start = time.time()
+    mdl = Model('VRPTW')
+
+    # Start time
+
+    e = [df['READYTIME'][i] for i in range(n + 1)]
+    e.append(df['READYTIME'][0])
+
+    # Due time
+
+    l = [df['DUETIME'][i] for i in range(n + 1)]
+    l.append(df['DUETIME'][0])
+
+    # Service time
+    ser = [df['SERVICETIME'][i] for i in range(n + 1)]
+    ser.append(df['SERVICETIME'][0])
+
+    # Demand
+
+    #r = {i: df['DEMAND'][i] for i in range(1, n + 1)}
+    r = [df['DEMAND'][i] for i in range(n + 1)]
+    r.append(0)
+
+    # Variable set
+    X = [(i, j, k) for i in Cc for j in Cc for k in V if i != j]
+    S = [(i, k) for i in Cc for k in V]
+
+    # Calculate distance and time
+    c = {(i, j): dist_matrix[i][j] for i in Cc for j in Cc}
+    t = {(i, j): dist_matrix[i][j] for i in Cc for j in Cc}
+
+    # Variables
+    x = mdl.binary_var_dict(X, name='x')
+    s = mdl.continuous_var_dict(S, name='s')
+
+    # Constraints
+    mdl.sum(c[i, j] * x[i, j, k] for i, j, k in X)
+
+    mdl.add_constraints(mdl.sum(x[i, j, k] for j in Cc for k in V if j != i) == 1 for i in C)
+
+    mdl.add_constraints(mdl.sum(r[i] * mdl.sum(x[i, j, k]) for i in C for j in Cc if i != j) <= Q for k in V)
+
+    mdl.add_constraints(mdl.sum(x[0, j, k] for j in Cc if j != 0) == 1 for k in V)
+
+    mdl.add_constraints(
+        (mdl.sum(x[i, p, k] for i in Cc if i != p) - mdl.sum(x[p, j, k] for j in Cc if p != j)) == 0 for p in C for k in
+        V)
+
+    mdl.add_constraints(mdl.sum(x[i, n + 1, k] for i in Cc if i != n + 1) == 1 for k in V)
+
+    mdl.add_constraints(s[i, k] + ser[i] + t[i, j] - K * (1 - x[i, j, k]) - s[j, k] <= 0 for i, j, k in X if i != j)
+
+    #mdl.add_constraints(s[i,k] + ser[i] + t[i,j] - (max(0, (l[i] + ser[i] + t[i,j] - e[j])))*(1-x[i,j,k]) - s[j,k] <=0 for i,j,k in X if i!=j if i!=n+1 if j!=0)
+
+    mdl.add_constraints(s[0, k] == 0 for k in V)
+
+    mdl.add_constraints(s[i, k] >= e[i] for i, k in S if i != 0)
+
+    mdl.add_constraints(s[i, k] <= l[i] for i, k in S if i != 0)
+
+    # Objective Function
+    obj_function = mdl.sum(c[i, j] * x[i, j, k] for i, j, k in X)
+
+    # Set time limit
+    mdl.parameters.timelimit.set(1000)
+    #mdl.parameters.emphasis.mip.set(3)
+    #mdl.parameters.mip.tolerances.mipgap.set(0.4)
+    #mdl.parameters.mip.strategy.probe.set(3)
+
+
+    # Solve
+    mdl.minimize(obj_function)
+
+    time_solve = time.time()
+
+    solution = mdl.solve(log_output = True)
+
+    time_end = time.time()
+    # print(solution)
+
+    running_time = round(time_end - time_solve, 2)
+    elapsed_time = round(time_end - time_start, 2)
+
+    if solution != None:
+        route = [x[0, i, k] for i in C for k in V if x[0, i, k].solution_value == 1]
+        no_vehicles = len(route)
+        obj = round(obj_function.solution_value, 2)
+        print(file, cust_size, no_vehicles, obj, elapsed_time, running_time)
+    else:
+        print(file, cust_size, 'NA', 'NA', elapsed_time, running_time)
+