Cinema screen plan with a Monte Carlo simulation

A group assignment. We simulated 100,000 nights of demand for two screen plans. The plan that we proposed gives more expected revenue from the same five screens.

Role
Group member (coursework)
Period
2022
Stack
Python, NumPy, Monte Carlo
Evidence
Modelled
Question
A cinema has five screens and three films. Which film goes on which screen, so that one night gives the most ticket revenue?
Data
The assumptions of the assignment brief: 440 seats on five screens, a mean of 400 guests in a night, a fixed share of guests for each film, and two ticket prices.
Method
We modelled the number of guests as a Poisson variable. We simulated 100,000 nights for the plan of the manager and 100,000 nights for our plan. For each night, the revenue is the seats that are sold times the ticket price.
ProjectionModelled
Expected revenue goes from about $5,756 to about $5,975 in a night. That is about $218 more from the same screens (modelled).

Model output from a group assignment under the assumptions of the brief. It is not a real cinema and not an operating result. The chance to reach the revenue target changes with the demand model, so this page does not give it.

100,000Simulated nights for each plan
$5,756Expected revenue, plan of the manager
$5,975Expected revenue, our plan
440Seats on five screens

1 · Business context

A cinema has five screens and shows three films on one night: Avatar, Black Adam and The Fabelmans. The manager must give each screen a film before she knows the demand.

She has a revenue target of $6,000 for the night.

2 · Simulation parameters

  • Guests in one night: a Poisson variable with a mean of 400.
  • Share of guests: Avatar 40%, Black Adam 40%, The Fabelmans 20%. That is about 160, 160 and 80 guests.
  • Prices: $19.99 for an IMAX seat and $12.99 for a regular seat.
  • Screens: A and B have 120 seats, C and D have 80 seats, E has 40 seats. The total is 440 seats.
  • 100,000 simulated nights for each plan.

3 · The plan of the manager

This plan gives 240 seats to Avatar for about 160 guests, and 40 seats to The Fabelmans for about 80 guests. About 357 of the 440 seats are filled on a mean night.

  • Screen A (120, IMAX): Avatar.
  • Screen B (120, IMAX): Avatar.
  • Screen C (80): Black Adam.
  • Screen D (80): Black Adam.
  • Screen E (40): The Fabelmans.

4 · Our plan

Avatar gets 200 seats, Black Adam gets 160 and The Fabelmans gets 80. About 395 of the 440 seats are filled on a mean night.

  • Screen A (120, IMAX): Avatar.
  • Screen B (120): Black Adam.
  • Screen C (80): Avatar.
  • Screen D (80): The Fabelmans.
  • Screen E (40): Black Adam.

5 · Result

Expected revenue goes from about $5,756 to about $5,975 in a night. The gain stays the same when the demand model changes, so it is the figure on this page.

Our plan also makes the $6,000 target more probable. The size of that change depends on how the demand for the three films is drawn, so this page does not give a value.

6 · Recommendations

  • Use the plan that fits seats to demand.
  • Do not give The Fabelmans only 40 seats. About 80 guests want it, so half of them find no seat.
  • Keep one Avatar screen in IMAX, because that seat has the highest price.
  • Use the same simulation when the films change. Only the demand shares need an update.

7 · Limits

  • The demand shares are fixed values from the brief.
  • A guest who finds no seat does not go to a different film.
  • The model has one night and one showing for each screen.
  • Prices do not change with demand.

Expected ticket revenue in one night

Expected ticket revenue in one night
ItemValue
Plan of the manager$5,756
Our plan$5,975
Our plan, as submitted (from the assignment notebook)
lam, sims = 400, 100000
exp_avatar = np.random.poisson(lam, sims) * 0.4
exp_black_adam = np.random.poisson(lam, sims) * 0.4
exp_the_fabelmans = np.random.poisson(lam, sims) * 0.2

screen_A = np.minimum(exp_avatar, cap_A)
screen_B = np.minimum(exp_black_adam, cap_B)
screen_C = np.minimum(exp_avatar - screen_A, cap_C)
screen_D = np.minimum(exp_the_fabelmans, cap_D)
screen_E = np.minimum(exp_black_adam - screen_B, cap_E)

total_exp_rev = screen_A * price_imax + (screen_B + screen_C + screen_D + screen_E) * price_reg
avg_exp_rev = np.mean(total_exp_rev)