Business Process Analysis and Simulation.

Health-Station Simulation

Your Advisory company has been
hired by HealthStation a 24-hour, 7 days a week clinic, to conduct a
simulation analysis of its operations. A week of patient interarrival time has
already been collected in the file:” patients IT” and must now be analyzed to
determine the appropriate statistical distribution to represent the patient
arrival patterns.
The clinic has five different
stations where patients might be directed for service. All patients begin at
the sign-in station staffed by a single receptionist. After sign-in, a patient
may be directed to either the Nurse’s Station (.9888 probability) or to an Exam
Room (0.0112 probability) to be seen by a doctor.
After the Nursing station the
patient is sent either to the Exam Rooms, with a probability of 90%, or to the
Lab/X-Ray with a probability of 10%.
After the Exam Rooms, a patient
has a 5% chance to be sent to the Nurse’s Station, 46% to the Exam Rooms, 3% to
Lab/X-Ray, and 46% to Check Out.
After the Lab/X-Ray, the patient
can be sent to the exam room with a probability of 83% or to the Check Out,
with a probability of 17%.
Note that if a patient is
directed to a repeat visit to the Exam-Room they are required to wait in line
again. All patients will check out when they are finished receiving medical
care and then leave the clinic. After the Check Out process, the patient leaves
the clinic.
The Sign-In station is staffed
by one receptionist who takes TRIA (1,4,8) (i.e a triangular distribution with
a minimum value of 1 minute, a most likely value of 4 minutes and a maximum
value of 8 minutes) to sign-in a patient. The Nurse’s Station is staffed by a
single nurse who takes TRIA (2,5,7) minutes to treat a patient. There are three
identical Exam Rooms, each staffed by a doctor who takes TRIA (4,7,13) minutes
to treat a patient. The Lab/X-Ray station can handle one patient at a time and
takes TRIA (15,23,38) minutes to process a patient. The Check Out station has
one person ensuring you pay before you leave, and this takes TRIA (3,5,8)
minutes.
HealthStation has experienced
low patient satisfaction (based on recent surveys), so they are looking to
increase patient (customer) satisfaction levels. We are evaluating possible
changes to the clinic, and you have been asked to provide an analysis and
recommendation.

Patients inter-arrival time
(minutes):
2.09
9.38
16.12
5.95
1.19
10.46
10.92
0.32
8.24
12.02
3.77
6.58
3.11
10.31
0.37
4.77
2.37
2.13
0.68
0.17
0.24
1.38
5.94
2.73
4.95
0.19
4.94
1.17
3.91
2.94
18.12
2.86
1.57
13.49
5.41
2.69
5.98
2.52
0.31
3.94
5.03
14.06
2.76
2.03
1.95
12.09
4.38
2.37
13.24
1.06

We have three options for
improving the operation of the clinic:

1) Add one additional nurse to
the Nurse’s station.

2) Invest in equipment that will
slightly speed up the exam room process, resulting in a new service time
of TRIA (3, 5, 12).

3) Perform an expanded set of
procedures in the exam room, resulting in a slightly longer service time (new
service time of TRIA(5, 9, 15) , but reducing the probability of a patient
being sent to another exam room by 10% (new probability is .3706) and increasing
the probability of a patient being sent to the check out by 10% (new
probability is .5652).

You are required to create an
initial model representing the current performance of the operations and
simulate it under the “baseline” (current) conditions, and then simulate it
with the optional changes. Assume all improvement options have the same
cost. Assume the clinic operates 24/7, and once a month it closes for a
while for maintenance, cleaning, and so on. So, please make your simulation
runtime (replication length) cover a one-month (30 day) time period.

The most important drivers of
patient satisfaction are the waiting times in the different stations, as well
as the total time spent in the clinic.
The goal of the project is
to:
–
Explain the current displeasure of patients.
–
Assess the effectiveness of the three options to improve the clinic.
– Choose one of the options and
explain why it is the best option to improve the clinic.

Submission file type:
1) A report that must follow the
specified format (description begins on the next page). Remember that what you
turn in should be a professional document. That means a single, well organized,
document in a pdf format.

2) A copy of your .mox
file (the Extendsim model file)

REPORT FORMAT FOR SIMULATION
PROJECT
–
Cover Page
–
Executive Summary
–
Main Report
– Appendices

Sections of the document should
be easily identified (i.e. labeled).
Some more details about
each section…
COVER PAGE
Please have a separate cover
page. The cover page should list:
•
Your name,
•
The title of the project (with graphics, if desired),
•
The date,
• The class.

EXECUTIVE SUMMARY
This is a very short synopsis of
the report, generally around 1/4 of a page (3-4 sentences). It should appear at
the beginning of the main report (not on a separate page). It gives a quick
overview of the report, including recommendations, and is not the same as the
introduction of your main report. It should answer the basic questions of “what
is the problem, what did you do, what did you find, and what do you recommend”?
This is used by upper management to get a quick idea of what is in the report
and to decide whether to have someone look more closely (i.e. read) the entire
report.
MAIN REPORT
This is a BUSINESS report, about
3-4 pages long. Keep the techy jargon to a minimum. The report should NOT be
about your model. Describe the problem, the important performance metrics, how
you approached the problem (simulation) and why simulation is the appropriate
choice, the results and analysis (only important points, summarized), and the
recommendation you are making based on your work. Technical details are left
for the appendices. Backup your recommendation with the results from your
simulation (that’s why you create the simulation model in the first place).
Don’t “oversell” your recommendation. Acknowledge that it may not be
the best solution for ALL situations (i.e. that tradeoffs exist) but that you
think that it will likely be the best course of action. Refer the reader to the
appropriate appendices to keep too much technical stuff from being in the main
report (i.e. give summarized info and tell the reader to see Appendix XX for
details).
This does not mean that no data
should appear in the report, just make sure it is important stuff. A summary
table or graph in the report can really help a non-technical reader understand
your point.
Using separate sections (that
are identified with headings) can make a report much easier to follow.
Here is a set of sections:
1)
Problem/Current situation (including the goal/decision(s) to make)
2)
Metrics Chosen (how to measure performance)
3)
Modeling Approach (briefly mention that you are using simulation modeling and
why that approach was chosen)
4)
Alternatives investigated (i.e. experimental design)
5)
Results
6) Recommended course of action

APPENDICES
These contain the technical work
you have performed and form a basis for the recommendations you have made. Not
everyone will read the Appendices, so you can assume the reader has more
technical knowledge and can present things with more jargon (try to define
terms that few people will know though).
Think about using bullets to
present the basic info, followed by a paragraph with more detail (if needed).
This is preferable to putting everything in paragraph format. This can be
useful in many sections such as the conceptual model, experimental design, and
so on.
Below are some comments on the
appendices:
Appendix A, Conceptual
Model (flow chart):
Use the flow chart
format/symbols we discussed in the lectures. Describe the situation in terms of
what we know about the current system, the problem, the decision variables
(things that we can change in the system, i.e. decisions to make), and the
performance metrics you are using to assess system performance.
The conceptual model should
describe the system in enough detail that someone who does not even know the
situation could develop a simulation model of the system based on the
conceptual model alone. Generally, we include some combination of flowcharts, structured
English (logic descriptions), listings of the important simulation pieces,
variables, statistics, and so on.
Appendix B, Extendsim
Model:
Describe how you took the
conceptual model and implemented it. Include the Extendsim flowchart (a
snapshot of your Extendsim model), and some text briefly explaining what
happens in important modules. Finally, be sure to describe any variables,
expressions, statistics, sets, and so on, which are used since they are not
obvious from looking at the flowchart.
Appendix C, Experimental
Design:
How you ran the model and how
you determined what to run. Includes scenarios (various levels of decision
variables) and run parameters (run length, # replications, etc).
Appendix D, Output: Include a summarized output for each scenario,
generally in table format. One table for each metric is usually appropriate for
a better comparison between the different scenarios. Remember, not every piece
of data needs to be included in the output summary- focus on the important
output (the things related to your metrics). Do not interpret the output, just
present it. Remember, confidence intervals (or the mean and half width) provide
much more useful information about the system than just point estimates (such
as the mean) do.
Appendix E, Analysis and
Sensitivity:
Here is where you interpret the
output (figure out what it means). You are attempting to show the differences
between choosing various courses of action, and the TRADEOFFS that are
involved. Only rarely will one course of action be the best in all measures of
performance. Statistical analysis of the results is an important part of this
section. Graphs can be very helpful in showing the relationship between a
performance measure and a decision variable.
Some comments:
• Please start each appendix
on a new page.

• Your report should be
consistent from section to section, including the appendices (ex. the sections
should not conflict with each other, it should be an integrated document not a
piecemeal collection of sections)

• Please use 1.5 line spacing
and reasonable font sizes and margins. Double-spacing, large font sizes, and
using large margins looks unprofessional.

• Avoid modeling-specific
terminology in the business report (“scenario”, “queue”, etc).

• Use descriptive names for your
scenarios in the report, not “option1”, “option2”, etc.

• The conceptual model and other
sections will benefit from using an outline format, bullet format, or some
other format that is easy to navigate (not big blocks of text). Everything does
not need to be a bullet, but using bullets and lists can make info easier to
find for the reader. Think about formatting so someone can scan an appendix and
find what they need quickly.

• The Modeling Approach
(methodology) section of the report is where you state that you are using
simulation modeling for the analysis and then justify why you are using
simulation. Information about what software you used is often included as well.

Last Completed Projects