2. Patients must be
addressed with the best
Emergency Departments (ED) quality.
are complex and
quite dynamic systems.
ED’s are overcrowded and work
with limited budget.
12. STATE Variables Values Observability
Name/identifier <id> Unique per agent I
Gender, Medical history (cardiology, pulmonology,
neurological,…); Allergies (yes-no);
Personal details Treatments that received (classified into therapeutic groups: I
bronchodilators, vasodilators, etc.);
Origin (national or immigrant)
Entrance, Admissions, Waiting Room, Triage, Treatment
Location E
Zone.
Idle, Requesting information from <id>, Giving information
Action to <id>, Searching, Moving to <location> , Waiting for E
ambulance.
Healthy; Hemodynamic-Constant; Barthel Index (degree of
Physical condition
Variables Values E/I/N
Observability
dependence).
Healthy, Cardiac/respiratory arrest, severe/moderate
Symptoms (patients) E/I
trauma, headache, vomiting, diarrhea
Communication skills Low, Medium, High E
Level of experience Resident (1 to 5); Junior (5-10); Senior (10 - 15) and E/I
Current state Next state / (doctors) Consultant (over 15 years)
Input
/ Output Output
…. …. …. Level of experience
E/I
(triage Low, Medium, High
Sx / Ox Ia (p1) Sy / Oy nurses)
Level of experience E/I
Sx / Ox Ia (p2) Sz / Oz (emergency nurses)
Low, Medium, High
Level of experience E/I
Sx / Ox Ia (p3) Sx / Ox (admissions)
Low, Medium, High
…. …. ….
13. 1) Active Agents
2) Passive Agents
Patients
Information system
Companions of patients
Admission personnel Loudspeaker system
Sanitarian technicians Pneumatic pipes
Nurses (Triage, Emergency) Tests services
Doctors (Emergency,
Specialists)
1 to Zone: individuals in Zone
1 to 1(One-to-One) 1 to n (Multicast)
(Area- Restricted Broadcast)
14. The Environment
Arrival/dismissal
by ambulance
Arrival/dismissal
by own means
The model should include the spatial topographical design from the ED
16. Arrival/dismissalb
y ambulance
Arrival/dismissal
by own means
A
N
D
17. ED Simulator
Input Patients arrival:
Could arrive every 3 min. , but with different probabilities:
20% (4 pat/hr), 40% (9 pat/hr),
60% (13 pat/hr) , 80% (17 pat/hr)
Configuration of the ED Staff
Staff Number Junior Senior
Admission 1-2 2 min. 1 min. 15 sec.
Triage Nurse 1-3 8 min. 5 min.
Doctor 1-4 20 min. 15 min.
Output Patients:
How many arrive to the service
How many leave the service
Times of staying in each area
What if?
18.
19. • Find the best/optimum solution from all the
possible solutions.
Given any objective (index) function f :
f :A
max / min f x
subject to x A
A constraintset; xo A
f xo f x f xo f x
Maximize minimize
xo A
20. Is it always the "best solution" (the
optimum) the most interesting for us?
21.
22. Methodology
Simulator: 2nd version
Parameter configuration: A, N, D = > 3D + P => 4D
A
N
D
~ 400 patients daily
23. Methodology: Computational complexity
Multidimensional Discrete
DD
D • Search space
– # Dimensions = Patients,
B
staff (D, N, A, …), T, B,
P
PP …
NN
N N
– Each dimension=>
range of possible
A
A
AA T values
– # Points = #
simulations
(indexes)(time)
COMBINATORIAL!
24. ABM SIMULATOR PARAMETERS
DSS I
N
D
E
+ X
constraints