Business Simulation Exam
nimab8
asg.pptxADM 3305 Business Simulation Analytics
Simulation of Stochastic Processes
The Appointment Scheduling Game
© 2020 Antoine Sauré
All forms (printed, digital, etc.) of course materials prepared by the instructor (including e-mailed or Brightspace content) are protected by copyright. This covers all files, assessments, solutions, cases, recordings and other materials. Copying, scanning, photographing, posting, or sharing by any means is a violation of copyright and will be subject to appropriate penalty as prescribed by University of Ottawa regulation.
1
THE APPOINTMENT SCHEDULING GAME
An Advance Appointment Scheduling Setting
It is the end of the day and requisitions for appointments that can be booked tomorrow or later are sitting in your inbox waiting to be assigned
There are different urgency categories for appointments, each category with a different wait time target
Assume that appointments can be scheduled on any day including the weekend (and holidays) and note that once an appointment day is set it cannot be changed
The Appointment Scheduling Game
Simulates a system in which daily patient appointment requests, which are characterized by their urgency level, arrive randomly and in which daily service capacity is limited
You will assume the role of a scheduling clerk who must assign appointment dates to these requests without knowing future demand
| Urgency category | Wait time target |
| 1 | 2 days |
| 2 | 4 days |
| 3 | 6 days |
WHITE x 1
| JULY 2012 | ||||||
| SUN | MON | TUE | WED | THU | FRI | SAT |
| 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 |
RED
x 2
BLUE
x 1
WHITE x 1
PATIENT ARRIVALS
3
2
3
3
2
3
3
3
3
3
2
3
1
2
Roll a die
The numbers indicate the total number of chips stacked on each day.
5
WHITE x 1
| JULY 2012 | ||||||
| SUN | MON | TUE | WED | THU | FRI | SAT |
| 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 |
5 PATIENT ARRIVALS
3
2
3
3
2
3
3
3
3
3
2
3
1
2
2
1
2
Reach into a sack and pick out as many chips as the number that appeared on the die
6
WHITE x 1
| JULY 2012 | ||||||
| SUN | MON | TUE | WED | THU | FRI | SAT |
| 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 |
BOOKING DECISIONS
3
2
3
3
2
3
3
3
3
3
2
3
1
2
2
1
2
Place the chips on the calendar without exceeding three chips—the daily capacity—on any day
Appointments can be scheduled on any calendar day.
students are not allowed to reschedule patients or postpone booking decisions to the next simulated day.
7
| JULY 2012 | ||||||
| SUN | MON | TUE | WED | THU | FRI | SAT |
| 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 |
3
2
3
3
2
3
3
3
3
3
3
3
3
2
2
Move to the next day and go to step 1
asg.sauder.ubc.ca
Game Settings
Settings:
ADM 3305 A: LA1BG
ADM 3305 B: xTjzJ
ADM 3305 C: ZnSnM
Current day (today)
Simulation length and number of days left
Daily capacity (appointment slots)
Current system workload (available appointment slots)
Patients waiting to be scheduled
Service metrics
System utilization
Discussion
What is realistic and what is unrealistic about the game setting?
In what type of situations is the type of scheduling considered by the game relevant?
What levers for managing capacity are available to regulate the scheduling process?
What are relevant performance metrics?
What information should a booking agent keep track of?
What scheduling rules would you use to book patient appointments?
How do these scheduling rules perform?
Markov Decision Processes
Sequential Decision Making Problems
At each decision time or stage, the decision maker observes the state of the system
Based on this information, the decision maker chooses an action from a set of alternatives
As a consequence, the decision maker receives and immediate reward (or incurs an immediate cost)
The system evolves to a new state influenced by the action taken (deterministically or probabilistically)
16
Sequential Decision Making Problems
| Element | Definition |
| Stage | Point in time at which a decision is made |
| Horizon | Number of stages |
| State | Description of the system that provides the decision maker with all of the information necessary to make future decisions |
| Action | One of several alternatives available to the decision maker when the system is observed in a particular state |
| Transition | Subsequent state conditional on the current state and the action taken |
| Reward | Immediate reward/cost associated with taking a particular action in a particular state |
17
Advance Appointment Scheduling
| Element | Definition |
| Stage | Days in the planning horizon |
| Horizon | Infinite |
| State(s) | The schedule that currently stands from today to the end of the booking horizon Demand waiting to be booked from each priority class |
| Action | Which available slots to assign to the incoming demand and the number of patients to serve through overtime |
| Transition(s) | Update of the number of available slots Next day’s incoming arrivals |
| Reward(s) | - Cost incurred if a patient if booked later than the recommended waiting time - Cost associated with using overtime |
Advance Appointment Scheduling
Start of Day n
Determine the number of available slots on each day in future
Day n demand arrives
Start of Day n+1
Assign priorities to each waiting request
Assign prioritized demand to available slots
| Challenge facing the resource manager |
| How to allocate available capacity to incoming demand so that waiting times are achieved in an efficient way? |
19
Simulation Approach
| Element | Definition |
| Decision rule | Function that determines for the decision maker which action to select in each possible state of the system |
| Policy | Sequence of decision rules |
s1
s2
sN
sN+1
s3
d1
d2
dN
d1(s1)
d2(s2)
dN(sN)
r1(s1, d1(s1))
r2(s2, d2(s2))
rN(sN, dN(sN))
1
2
N
Decision Rule
State
Reward
Stage
20
21
