def run_simulation(run, value, value_DOW, value_month):
###parameters
set_price = 71
gamma = 20
average_number_of_potential_customers = 100
min_budget = 30
capacity = 76
max_lead = 200
###Weights
month_prob = [0.75, 0.8, 0.95, 0.9, 1., 0.95, 0.95, 1., 1., 1., 1.1, 1.2]
dow_prob = [1, 1, 1, 1, 1, 1, 0.7]
gamma_weights = [
1.1, 0.8, 1.2, 1.1, 0.6, 1.1, 1.2, 0.9, 0.7, 1.0, 0.9, 0.9
]
gamma_weights_dow = [1.05, 1.05, 1.05, 1.05, 0.9, 0.95, 0.8]
###Start/End date
start_reserved_night_date = dt.datetime(2018, 1, 1)
end_reserved_night_date = dt.datetime(2018, 12, 30)
###Define Hotel. (Doesn't do much now, but in future will have multiple merchants/room types)
hotel = Hotel(merchant_id=1,
capacities={10: capacity},
available_rooms=[10])
#Initial prices for each DOW
prices = {
10: {
0: set_price,
1: set_price,
2: set_price,
3: set_price,
4: set_price,
5: set_price,
6: set_price
},
}
###Open and write first line of output files
"""with open("transformed_data.csv", "w") as data:
writer = csv.writer(data, delimiter=',', quotechar='"', quoting=csv.QUOTE_MINIMAL)
writer.writerow(['hotelroom_id','reserved_night_date','booking_datetime','type','price'])
person_data=open("person_data.csv", "w")
pers_writer = csv.writer(person_data, delimiter=',', quotechar='"',quoting=csv.QUOTE_MINIMAL)
pers_writer.writerow(['booking_datetime','price'])"""
###Call revenue manager (currently very simple)
lead_rm = RLRevenueManager(
prices=prices,
hotel=hotel,
start_reserved_night_date=start_reserved_night_date,
end_reserved_night_date=end_reserved_night_date)
start_date = start_reserved_night_date
end_date = end_reserved_night_date
### initialise number of customers, budget and interested date
market = Market(start_reserved_night_date=start_reserved_night_date,
end_reserved_night_date=end_reserved_night_date,
average_number_of_potential_customers=
average_number_of_potential_customers,
min_budget=min_budget,
gamma=gamma,
gamma_weights_months=gamma_weights,
gamma_weights_dow=gamma_weights_dow,
month_prob=month_prob,
dow_prob=dow_prob,
capacity=capacity,
state=value,
value_DOW=value_DOW,
value_month=value_month)
#market.initialize_market()
### iterate over days
maximum_wtps = {}
for delta in range((end_date - start_date).days):
reserved_night_date = start_date + dt.timedelta(days=delta)
#print("Processing", reserved_night_date)
person_np = market._create_persons(reserved_night_date) #,pers_writer)
#person_np[:,6]==delta
#print(person_np[:,6],delta)
lead_rm = market.generate_bookings_for_reserved_night_date(
reserved_night_date, lead_rm, person_np, run)
print('Total Visits', np.sum(market.state[:, 1]))
#More output writing
"""bookings= np.array([[0,dt.datetime(2018,1,1),0]]*len(lead_rm._current_bookings))
bookings[:,0] = lead_rm._current_bookings[:,2]
bookings[:,1] = lead_rm._current_bookings[:,0]
bookings[:,2] = lead_rm._current_bookings[:,1]
df= pd.DataFrame(data=bookings)
df.columns=['hotelroom_id','reserved_night_date','units']
df.to_csv('transformed_room_nights.csv',index=False)
person_data.close()"""
return lead_rm._revenue, market.state, market.value_DOW, market.value_month
def run_simulation():
###parameters
set_price=71 # sets average price shown by revenue manager
gamma = 20 # sets width of booking time distribution
average_number_of_potential_customers = 100
min_budget=30
capacity = 76 # No. of rooms in hotel
###Weights of price for month and day of week variation
month_prob = [0.75,0.8,0.95,0.9,1.,0.95,0.95,1.,1.,1.,1.1,1.2]
dow_prob = [1,1,1,1,1,1,0.7]
gamma_weights = [1.1,0.8,1.2,1.1,0.6,1.1,1.2,0.9,0.7,1.0,0.9,0.9]
gamma_weights_dow = [1.05,1.05,1.05,1.05,0.9,0.95,0.8]
###Start/End date
start_reserved_night_date = dt.datetime(2018, 1, 1)
end_reserved_night_date = dt.datetime(2018, 12, 30)
###Define Hotel. (Doesn't do much now, but in future will have multiple merchants/room types)
hotel = Hotel(merchant_id=1, capacities={10: capacity}, available_rooms=[10])
#Initial prices for each DOW
prices = {
10: {0: set_price, 1: set_price, 2: set_price, 3: set_price, 4: set_price, 5: set_price, 6: set_price},
}
###Open and write first line of output files
with open("transformed_data.csv", "w") as data:
writer = csv.writer(data, delimiter=',', quotechar='"', quoting=csv.QUOTE_MINIMAL)
writer.writerow(['hotelroom_id','reserved_night_date','booking_datetime','type','price'])
person_data=open("person_data.csv", "w")
pers_writer = csv.writer(person_data, delimiter=',', quotechar='"',quoting=csv.QUOTE_MINIMAL)
pers_writer.writerow(['booking_datetime','price'])
###Call revenue manager (currently very simple human like)
lead_rm = LeadRevenueManager(prices=prices, hotel=hotel, start_reserved_night_date=start_reserved_night_date,
end_reserved_night_date=end_reserved_night_date)
start_date = start_reserved_night_date - dt.timedelta(days=gamma * 10)
end_date = end_reserved_night_date
### initialise number of customers, budget and interested date
market = Market(start_date=start_date,
start_reserved_night_date=start_reserved_night_date,
end_reserved_night_date=end_reserved_night_date,
average_number_of_potential_customers=average_number_of_potential_customers,
minimum_budget_in_population=min_budget,gamma=gamma,gamma_weights_months=gamma_weights,
gamma_weights_dow=gamma_weights_dow,month_prob=month_prob,dow_prob=dow_prob,capacity=capacity
)
### iterate over lead time
for delta in range((end_date-start_date).days):
time = start_date + dt.timedelta(days=delta)
#print("Processing",time)
RND = start_reserved_night_date + dt.timedelta(days=delta)
if RND<end_date:
person_np = market._create_persons(RND,pers_writer)
if delta == 0:
person = person_np
else:
person = np.vstack([person,person_np])
where = (person[:,6]==delta)
lead_rm = market.generate_bookings_for_lead_time(lead_rm,person,time,where)
inv_where=np.invert(where)
person = person[inv_where]
###Output
bookings= np.array([[0,dt.datetime(2018,1,1),0]]*len(lead_rm._current_bookings))
bookings[:,0] = lead_rm._current_bookings[:,2]
bookings[:,1] = lead_rm._current_bookings[:,0]
bookings[:,2] = lead_rm._current_bookings[:,1]
df= pd.DataFrame(data=bookings)
df.columns=['hotelroom_id','reserved_night_date','units']
df.to_csv('transformed_room_nights.csv',index=False)
person_data.close()
### Return total revenue
return lead_rm._revenue
