-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathDispatcher.py
522 lines (418 loc) · 23.5 KB
/
Dispatcher.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
import random as r
import matplotlib.pyplot as plt
import math
from Dijkstra import *
import numpy as np
import time
from optim import opt
from scipy.spatial import Delaunay
from insertion import *
class graph:
def __init__(self, size, size_x, size_y):
self.size = size #int
self.size_x = size_x #m
self.size_y = size_y #m
self.initialise_nodes()
self.initialse_arcs()
def initialise_nodes(self):
self.nodes = []
for i in range(self.size):
x = self.size_x*r.random()
y = self.size_y*r.random()
self.nodes.append(node(len(self.nodes)+1,x,y))
def initialse_arcs(self):
self.arcs = []
points = np.array([[node.x, node.y] for node in self.nodes])
tri = Delaunay(points)
tmp_dict = {i:set() for i in range(len(self.nodes))}
for simp in tri.simplices:
tmp_dict[simp[0]].add(simp[1])
tmp_dict[simp[0]].add(simp[2])
tmp_dict[simp[1]].add(simp[0])
tmp_dict[simp[1]].add(simp[2])
tmp_dict[simp[2]].add(simp[0])
tmp_dict[simp[2]].add(simp[1])
for i in range(len(self.nodes)):
tmp_dict[i] = list(tmp_dict[i])
for j in range(len(tmp_dict[i])):
self.arcs.append(arc(len(self.arcs)+1,self.nodes[i],self.nodes[tmp_dict[i][j]]))
def plot(self):
plt.figure(num=None, figsize=(10, 10), dpi=120, facecolor='w', edgecolor='w')
plt.plot([node.x for node in self.nodes], [node.y for node in self.nodes],'ro')
for arc in self.arcs:
plt.plot([arc.start_node.x, arc.end_node.x], [arc.start_node.y, arc.end_node.y])
plt.show
class node:
def __init__(self,ID,x,y):
self.ID = ID
self.x = x
self.y = y
def __str__(self):
return "Node: ID:%i, x:%f, y:%f" % (self.ID, self.x, self.y)
def __repr__(self):
return "(%i,%.2f,%.2f)" % (self.ID, self.x, self.y)
class arc:
def __init__(self,ID,start_node,end_node):
self.ID = ID
self.start_node = start_node
self.end_node = end_node
self.length = math.ceil(math.sqrt(pow(start_node.x-end_node.x,2)+pow(start_node.y-end_node.y,2)))
def __str__(self):
return "Arc: ID:%i, s:%i, e:%i" % (self.ID, self.start_node.ID, self.end_node.ID)
def __repr__(self):
return "(%i,%i,%i)" % (self.ID, self.start_node.ID, self.end_node.ID)
class user:
def __init__(self, ID, start_time, start_node, end_node):
self.ID = ID
self.start_time = start_time
self.start_node = start_node
self.end_node = end_node
def __repr__(self):
return "user-%i" % self.ID
class vehicle:
def __init__(self, ID, start_node, capacity):
self.ID = ID
self.node = start_node
self.capacity = capacity
self.path = []
self.short_path = []
self.users = []
self.target_users = []
self.arrival_time = 1
def __repr__(self):
return "vehicle-%i" % self.ID
def pickup(self, user, time):
if (not self.node == user.start_node) or len(self.users) == self.capacity:
raise Exception("Disallowed pickup")
elif time < user.start_time:
return "wait"
else:
self.target_users.remove(user)
self.users.append(user)
return "pickup"
def dropoff(self, user):
if (not self.node == user.end_node) or not user in self.users:
raise Exception("Disallowed dropoff")
else:
self.users.remove(user)
def assign(self, user):
self.target_users.append(user)
def deassign(self, user):
if user in self.target_users:
self.target_users.remove(user)
else:
raise Exception("%s is not in %s and can't be deassigned" % (user, self))
class model:
def __init__(self, size, size_x, size_y, n_users, n_vehicles, vehicle_cap, time, reopt, horizon):
self.graph = graph(size, size_x, size_y)
self.create_user_template(n_users, time)
self.create_vehicle_template(n_vehicles, vehicle_cap)
self.time = time
self.vehicle_cap = vehicle_cap
self.reopt = reopt
self.horizon = horizon
self.dist_dict = {}
self.path_dict = {}
for node1 in self.graph.nodes:
for node2 in self.graph.nodes:
self.path_dict[(node1, node2)], self.dist_dict[(node1, node2)] = dijkstra(node1, node2, self.graph.nodes, self.graph.arcs)
def create_user_template(self, n_users, time):
self.users_template = []
for i in range(n_users):
start_node = r.sample(self.graph.nodes,1)[0]
end_node = r.sample([node for node in self.graph.nodes if node != start_node],1)[0]
start_time = r.randint(1,math.ceil((1/2)*time))
self.users_template.append(user(len(self.users_template)+1, start_time, start_node, end_node))
def read_user_template(self):
tmp_users = []
for u in self.users_template:
tmp_users.append(user(u.ID, u.start_time, u.start_node, u.end_node))
return tmp_users
def create_vehicle_template(self, n_vehicles, vehicle_cap):
self.vehicles_template = []
for i in range(n_vehicles):
start_node = r.sample(self.graph.nodes,1)[0]
self.vehicles_template.append(vehicle(len(self.vehicles_template)+1, start_node, vehicle_cap))
def read_vehicle_template(self):
tmp_vehicles = []
for v in self.vehicles_template:
tmp_vehicles.append(vehicle(v.ID, v.node, v.capacity))
return tmp_vehicles
def simulate(self, dispatcher, mode):
start_time = time.time()
self.unserved_users = []
self.n_served = 0
self.users = self.read_user_template()
self.vehicles = self.read_vehicle_template()
self.waiting_times_dict = {user:np.inf for user in self.users}
time_user_dict = {t:[] for t in range(1,self.time+1)}
if mode[0] in ['dynamic', 'dynamic_insert']:
for user in self.users:
time_user_dict[user.start_time].append(user)
elif mode[0] == 'static':
for user in self.users:
time_user_dict[1].append(user)
self.at_node = {t:[] for t in range(1,self.time+1)}
for vehicle in self.vehicles:
self.at_node[vehicle.arrival_time].append(vehicle)
pickup_times = {v:[] for v in self.vehicles}
for t in range(1,self.time):
# ADD USERS
self.unserved_users.extend(time_user_dict[t])
# VEHICLE ARRIVALS
for vehicle in self.at_node[t]:
if vehicle.path:
vehicle.node = vehicle.path[0]
vehicle.path = vehicle.path[1:]
# onNextTimestep
dispatcher.on_next_timestep(self, t, self.vehicle_cap, mode, time_user_dict[t])
for vehicle in self.at_node[t]:
cont = True
while cont:
cont = False
for user in vehicle.users:
if user.end_node == vehicle.node == vehicle.short_path[0][0]:
assert vehicle.node == vehicle.short_path[0][0], "%s : %s : %s : %s" % (vehicle, user, vehicle.node, vehicle.short_path[0])
vehicle.short_path = vehicle.short_path[1:]
pickup_times[vehicle].append(t)
# print("%s dropoff %s at %i" % (vehicle, user, t))
vehicle.dropoff(user)
dispatcher.on_dropoff(self, vehicle)
self.n_served = self.n_served + 1
self.waiting_times_dict[user] = t
cont = True
break
wait = False
cont = True
while cont:
cont = False
if vehicle.target_users:
user = vehicle.target_users[0]
if user.start_node == vehicle.node == vehicle.short_path[0][0]:
res = vehicle.pickup(user, t)
if res == "wait":
vehicle.path = [vehicle.node] + vehicle.path
wait = True
elif res == "pickup":
# print("%s pickup %s at %i" % (vehicle, user, t))
# print(vehicle.node, user.start_node)
assert vehicle.node == vehicle.short_path[0][0], "%s : %s : %s : %s" % (vehicle, user, vehicle.node, vehicle.short_path[0])
vehicle.short_path = vehicle.short_path[1:]
self.unserved_users.remove(user)
dispatcher.on_pickup(self, vehicle)
pickup_times[vehicle].append(t)
cont = True
if not wait and vehicle.path:
if math.ceil(self.dist_dict[(vehicle.node, vehicle.path[0])]) > 0:
if t+math.ceil(self.dist_dict[(vehicle.node, vehicle.path[0])]) <= self.time:
self.at_node[t+math.ceil(self.dist_dict[(vehicle.node, vehicle.path[0])])].append(vehicle)
vehicle.arrival_time = t+math.ceil(self.dist_dict[(vehicle.node, vehicle.path[0])])
else:
pass
else:
assert vehicle.node == vehicle.path[0]
vehicle.node = vehicle.path[0]
vehicle.path = vehicle.path[1:]
if vehicle.path:
assert math.ceil(self.dist_dict[(vehicle.node, vehicle.path[0])]) > 0
self.at_node[t+math.ceil(self.dist_dict[(vehicle.node, vehicle.path[0])])].append(vehicle)
vehicle.arrival_time = t+math.ceil(self.dist_dict[(vehicle.node, vehicle.path[0])])
else:
self.at_node[t+1].append(vehicle)
vehicle.arrival_time = t+1
else:
self.at_node[t+1].append(vehicle)
vehicle.arrival_time = t+1
self.at_node[t] = []
if time.time()-start_time >300:
return [-1,-1]
for t in range(1,self.time):
for user in time_user_dict[t]:
if self.waiting_times_dict[user] == np.inf:
self.waiting_times_dict[user] = self.time-user.start_time
else:
self.waiting_times_dict[user] = self.waiting_times_dict[user]-user.start_time
print(time.time()-start_time)
return [sum([x for x in self.waiting_times_dict.values()]), time.time()-start_time]
class final_dispatcher():
def __init__(self):
self.to_update = []
def on_next_timestep(self, model, t, Q_max, mode, new_users):
err_penalty = 2000
n_iter = 250
if mode[0] == 'static':
freq = 10000
T = model.time
d_val = model.time
elif mode[0] in ['dynamic', 'dynamic_insert']:
freq = model.reopt
T = min(model.time-t, model.horizon)
d_val = T
if t % freq == 1:
print("t: %i" % t)
passenger_dict = {} # vehicle index : passenger index
passengers = []
arrival_dict = {c:0 for c in range(len(model.vehicles))}
tmp_passenger_no = 0
keep_track_of_passengers_dict = {} # passenger index : user
for vehicle in model.vehicles:
vehicle.target_users = []
passenger_dict[model.vehicles.index(vehicle)] = []
for user in vehicle.users:
passenger_dict[model.vehicles.index(vehicle)].append(tmp_passenger_no)
keep_track_of_passengers_dict[tmp_passenger_no] = user
tmp_passenger_no = tmp_passenger_no + 1
passengers.extend(vehicle.users)
l = len(passengers)
n = len(model.unserved_users)
m = len(model.vehicles)
n_nodes = 2*n + m + l + 1
travel_times = {("n%i" % (i+1), "n%i" % (j+1)):1000 for i in range(n_nodes) for j in range(n_nodes)}
for user1 in model.unserved_users:
for user2 in model.unserved_users:
travel_times[("n%i" % (model.unserved_users.index(user1) + 1), "n%i" % (model.unserved_users.index(user2) + 1))] = model.dist_dict[(user1.start_node, user2.start_node)]
travel_times[("n%i" % (model.unserved_users.index(user1) + n + 1), "n%i" % (model.unserved_users.index(user2) + 1))] = model.dist_dict[(user1.end_node, user2.start_node)]
travel_times[("n%i" % (model.unserved_users.index(user1) + 1), "n%i" % (model.unserved_users.index(user2) + n + 1))] = model.dist_dict[(user1.start_node, user2.end_node)]
travel_times[("n%i" % (model.unserved_users.index(user1) + n + 1), "n%i" % (model.unserved_users.index(user2) + n + 1))] = model.dist_dict[(user1.end_node, user2.end_node)]
for user2 in passengers:
travel_times[("n%i" % (model.unserved_users.index(user1) + 1), "n%i" % (2*n + m + passengers.index(user2) + 1))] = model.dist_dict[(user1.start_node, user2.end_node)]
travel_times[("n%i" % (model.unserved_users.index(user1) + n + 1), "n%i" % (2*n + m + passengers.index(user2) + 1))] = model.dist_dict[(user1.end_node, user2.end_node)]
travel_times[("n%i" % (model.unserved_users.index(user1) + 1), "n%i" % (2*n+m+l+1))] = 0
travel_times[("n%i" % (model.unserved_users.index(user1) + n + 1), "n%i" % (2*n+m+l+1))] = 0
for user1 in passengers:
for user2 in model.unserved_users:
travel_times[("n%i" % (2*n + m + passengers.index(user1) + 1), "n%i" % (model.unserved_users.index(user2) + 1))] = model.dist_dict[(user1.end_node, user2.start_node)]
travel_times[("n%i" % (2*n + m + passengers.index(user1) + 1), "n%i" % (model.unserved_users.index(user2) + n + 1))] = model.dist_dict[(user1.end_node, user2.end_node)]
for user2 in passengers:
travel_times[("n%i" % (2*n + m + passengers.index(user1) + 1), "n%i" % (2*n + m + passengers.index(user2) + 1))] = model.dist_dict[(user1.end_node, user2.end_node)]
travel_times[("n%i" % (2*n + m + passengers.index(user1) + 1), "n%i" % (2*n+m+l+1))] = 0
for vehicle in model.vehicles:
if vehicle in model.at_node[t]:
node = vehicle.node
else:
node = vehicle.path[0]
for user2 in model.unserved_users:
travel_times[("n%i" % (2*n + model.vehicles.index(vehicle) + 1), "n%i" % (model.unserved_users.index(user2) + 1))] = model.dist_dict[(node, user2.start_node)]
travel_times[("n%i" % (2*n + model.vehicles.index(vehicle) + 1), "n%i" % (model.unserved_users.index(user2) + n + 1))] = model.dist_dict[(node, user2.end_node)]
for user2 in passengers:
travel_times[("n%i" % (2*n + model.vehicles.index(vehicle) + 1), "n%i" % (2*n + m + passengers.index(user2) + 1))] = model.dist_dict[(node, user2.end_node)]
travel_times[("n%i" % (2*n + model.vehicles.index(vehicle) + 1), "n%i" % (2*n+m+l+1))] = 0
for node in range(n_nodes):
travel_times[("n%i" % (node+1), "n%i" % (node+1))] = 1000
for tt in range(t, min(model.time, t+T+1)):
for vehicle in model.at_node[tt]:
arrival_dict[model.vehicles.index(vehicle)] = tt-t
starting_times = {("n%i" % (i + 1)):(max(0,model.unserved_users[i].start_time-t)) for i in range(n)}
hot_start_paths = {}
hot_start_costs = {}
for vehicle in model.vehicles:
hs_cost = 0
if vehicle in model.at_node[t]:
last_node = vehicle.node
else:
last_node = vehicle.path[0]
load = len(vehicle.users)
time = arrival_dict[model.vehicles.index(vehicle)]
hs_path = ["n%i" % (2*n + model.vehicles.index(vehicle) + 1)]
p_users = []
for node in vehicle.short_path:
node = node[0]
time = time + model.dist_dict[(last_node, node)]
last_node = node
for user in vehicle.users:
if node == user.end_node:
hs_path.append("n%i" % (2*n + m + passengers.index(user) + 1))
hs_cost = hs_cost + time
load = load - 1
for user in vehicle.target_users:
if node == user.start_node:
hs_path.append("n%i" % (model.unserved_users.index(user) + 1))
if time < starting_times["n%i" % (model.unserved_users.index(user) + 1)]:
time = starting_times["n%i" % (model.unserved_users.index(user) + 1)]
p_users.append(user)
load = load + 1
if node == user.end_node:
hs_path.append("n%i" % (model.unserved_users.index(user) + n + 1))
hs_cost = hs_cost + time - starting_times["n%i" % (model.unserved_users.index(user) + 1)]
assert user in p_users
load = load - 1
assert load <= model.vehicle_cap
hot_start_paths[model.vehicles.index(vehicle)] = hs_path + ["n%i" % (2*n+m+l+1)]
hot_start_costs[model.vehicles.index(vehicle)] = hs_cost
ret_dict, MP, LP = opt(travel_times, m, n, l, passenger_dict, arrival_dict, starting_times, hot_start_paths, hot_start_costs, d_val, T, Q_max, err_penalty, n_iter, mode[1])
duplicate_list = []
for i in range(len(ret_dict)):
assert ret_dict[i+1][0] == "n%i" % (2*n + i + 1)
tmp_path = [int(x[1:])-1 for x in ret_dict[i+1]]
remove_indices = []
for j in range(len(tmp_path)):
if tmp_path[j] in duplicate_list:
remove_indices.append(j)
else:
duplicate_list.append(tmp_path[j])
remove_indices.reverse()
for index in remove_indices:
tmp_path.pop(index)
vehicle = model.vehicles[i]
if vehicle in model.at_node[t]:
next_node = vehicle.node
else:
next_node = vehicle.path[0]
vehicle.path = []
vehicle.short_path = []
for j in tmp_path:
if j<n:
vehicle.target_users.append(model.unserved_users[j])
vehicle.path.extend(model.path_dict[(vehicle.path[-1], model.unserved_users[j].start_node)])
vehicle.short_path.append((model.unserved_users[j].start_node, 's'))
elif n<=j<2*n:
vehicle.path.extend(model.path_dict[(vehicle.path[-1], model.unserved_users[j-n].end_node)])
vehicle.short_path.append((model.unserved_users[j-n].end_node, 'e'))
elif 2*n<=j<2*n+m:
assert vehicle.path == []
vehicle.path.append(next_node)
elif 2*n+m<=j<2*n+m+l:
vehicle.path.extend(model.path_dict[(vehicle.path[-1], keep_track_of_passengers_dict[(j-2*n-m)].end_node)])
vehicle.short_path.append((keep_track_of_passengers_dict[(j-2*n-m)].end_node, 'e'))
elif mode[0] == 'dynamic_insert':
for user in new_users:
insert(user, model, t)
def on_pickup(self, model, vehicle):
pass
def on_dropoff(self, model, vehicle):
pass
#D = 5
#n = [60]*D + [60]*D + [80]*D + [40]*D + [80]*D
#m = [5]*D + [5]*D + [10]*D + [5]*D + [10]*D
#t = [800]*D + [1000]*D + [800]*D + [600]*D + [800]*D
#cap = [2]*D + [2]*D + [2]*D + [2]*D + [2]*D
#p = [30]*D + [30]*D + [30]*D + [30]*D + [30]*D
#reopt = [50]*D + [50]*D + [50]*D + [50]*D + [30]*D
#horizon = [200]*D + [200]*D + [300]*D + [200]*D + [200]*D
#seeds = [i+j for i in [12314,34634,214,3245,1516] for j in [124,435,3,2357,137,1,2345,5754,7,12346,45673,1345,2135752,2342,5834,863,242342,23525,26264567,1324325,124235,1452362,15346,2347,564,124234,23214,4363456,1234,124]]*5
#niter = 5
#F = open('gen6.txt','w')
#time1 = []
#time2 = []
#val1 = []
#val2 = []
#for i in range(len(n)):
# print(i)
# r.seed(seeds[i])
# mod = model(p[i],100,100,n[i],m[i],cap[i],t[i], reopt[i], horizon[i])
# d = final_dispatcher()
# dyn = mod.simulate(d, ('dynamic_insert','label'))
# dyni = mod.simulate(d, ('dynamic_insert','reinsert'))
# both = mod.simulate(d, ('dynamic_insert', 'both'))
# F.write("&%.1f & %.1f && %.1f & %.1f && %.1f & %.1f\n" % (dyn[0], dyn[1],dyni[0], dyni[1], both[0], both[1]))
# val1.append(dyn[0])
# val2.append(dyni[0])
# time1.append(dyn[1])
# time2.append(dyni[1])
#F.close()
#
#vall1 = [sum(val1[:D])/D, sum(val1[150:300])/150, sum(val1[300:450])/150, sum(val1[450:600])/150, sum(val1[600:750])/150]
#vall2 = [sum(val2[:D])/D, sum(val2[150:300])/150, sum(val2[300:450])/150, sum(val2[450:600])/150, sum(val2[600:750])/150]
#timee1 = [sum(time1[:D])/D, sum(time1[150:300])/150, sum(time1[300:450])/150, sum(time1[450:600])/150, sum(time1[600:750])/150]
#timee2 = [sum(time2[:D])/D, sum(time2[150:300])/150, sum(time2[300:450])/150, sum(time2[450:600])/150, sum(time2[600:750])/150]
# &2119.0 & 1.2 & 2135.0 & 0.9 & 1487.0 & 3.7