You are on page 1of 7

REG NO 10045

Case Study
Quantitative Methods for Decision Making

Saad Khawar
1/3/2010

The document includes solution to the case study given in class MBA-1 in first semester course
QMDM taught by Dr. Ahmed Ali Shah
Submitted to: Dr. Ahmed Ali Shah
Saad Khawar – Case Study - I - QMDM 2

1. Write summary of given and the result required?

It is an application of linear programming used to measure the relative efficiency of operating units with
the same goals and objectives. This case is about the use Data Envelopment Analysis to measure the
relative efficiency of County hospital. For this purpose we have used a linear programming model to
construct a hypothetical composite hospital based on the outputs and inputs for the four hospitals in
the problem. For each operating unit that we want to measure the efficiency of, we must formulate and
solve a linear programming model similar to the linear program we solved to measure the relative
efficiency of County Hospital. The following step by step procedure should help you in formulating a
linear programming model for other types of DEA applications. Note that the operating unit we want to
measure the relative efficiency of is referred to as the jth operating unit.
Saad Khawar – Case Study - I - QMDM 3

2. The theory used in setting the equations

A linear programming model is developed for each hospital whose efficiency has to be evaluated. Then a
composite hospital is created based on the outputs and inputs for all the four hospitals.

Output Measures
 Patient-days of service under Medicare
 Patient-days of service not under Medicare
 Number of nurses trained
 Number of interns trained

Input Measures
 The number of full-time equivalent (FTE) non physician personnel
 The amount spent on supplies
 The number of bed-days available
 

Variables to be determined

wg weight applied to inputs and outputs for General Hospital

wu weight applied to inputs and outputs for University Hospital

wc weight applied to inputs and outputs for County Hospital

ws weight applied to inputs and outputs for State Hospital

Constraints

The DEA approach requires that the sum of these weights equal 1. Thus, the first constraint is
wg + wu + wc + ws = 1

Output Constraints

Medicare Patient-Days for = (Medicare Patient-Days for General Hospital)*wg +


composite hospital
(Medicare Patient-Days for County Hospital)*wc +
(Medicare Patient-Days for University hospital)*wu +
(Medicare Patient-Days for state hospital)*ws +
Saad Khawar – Case Study - I - QMDM 4

Medicare patient-days
= 48.14wg + 34.62wu + 36.72wc + 33.16ws
for Composite Hospital

The other output measures for the composite hospital are computed in a similar fashion.

General Format for the output Output for the composite hospital >= Output for county hospital
constraints
Medicare Patient days 48.14 wg + 34.62 wu + 36.72 wc + 33.16 ws >= 36.72
Non medicare patient days 43.10 wg + 27.11 wu + 45.98 wc + 56.46 ws >= 45.98
Nurses 253 wg+148 wu+175 wc+l60 ws >= 175
Interns 41 wg+27 wu+23 wc+84 ws >= 23

Input Constraints

(FTE nonphysicians
(FTE nonphysicians
FTE nonphysicians = for General Hospital)wg + for University Hospital)wu
for Composite Hospital
(FTE nonphysicians (FTE nonphysicians
+ County Hospital)wc
for + for State Hospital)ws

The other input measures for the composite hospital are computed in a similar fashion.

General Format for the input Output for the composite hospital >= Output for county hospital
constraints
FTE physicians 285.20wg + 162.30wu + 275.70wc + 210.40ws
Supple Expense 123.80wg + 128.70wu + 348.50wc + 154.10ws
Bed-days 106.70wg + 64.21wu + 104.10wc + 104.04ws

Objective Function

Minimize E = the fraction of County Hospital’s input available to the composite hospital
E is also known as efficiency index

The input constraint corresponding to FTE non physicians, supplies and bed-days:
285.20wg + 162.30wu +275.70wc+210.40ws <=275.70E (FTE non-physicians)
123.80wg + 128.70wu + 348.50wc + 154.10ws <= 348.50E (Supplies)
106.72wg + 64.21wu + 104.10wc ± 104.04ws <= 104.10E (Bed-days)

The complete model


wg + wu + wc + ws = 1
Saad Khawar – Case Study - I - QMDM 5

48.14wg+ 34.62wu +36.72wc+33.16ws >=36.72


43.10wg+ 27.11wu +45.98wc+33.16ws >=45.98
253wg + 148wu +175wc + 160ws >=175
41wg + 27wu +23wc + 84ws >=23
123.80wg+ 128.70wu +348.50wc+154.10ws <=348.50E
285.20wg+ 162.30wu +275.70wc+210.40ws <=275050E
106.72wg + 64.21wu + 104.10wc + 104.04ws<= 104.10E
E, wg, wu, ws, wc >= 0

3. Technique used to solve the equations with the proof

The equations have been solved using the Solver tool in the Microsoft Excel 2007. The screenshot has
been attached for further reference. The solver utility requires us to give the values of the constraints
and the input values. The computation is done automatically and the result is produced. It is a very
useful utility and helps in determining the values of n number of variables.

In the solver application tool in Excel 2007, the following steps are involved in getting to the end
required result.

1. The ‘set target cell’ represents the cell where you want the value of the objective function. The
‘equal to’ value is set to Max, Min or certain value of. In this case since E is a min function I have
set it to ‘Min’.
2. The ‘by changing cells’ identifies those cells for which the variable values are to be posted.
3. In the ‘subject to the constraint’ area the constraints for the inequalities are given.
4. When all the information is entered the ‘solver’ button is pressed which gives the result on the
excel sheet. The results are given as follows.

Hospital Weightage variable Value


General Hospital Wg 0.212
University Hospital Wu 0.260
State Hospital Ws 0.527
County Hospital Wc 0.000

s.no Constraints Limit Values Slack


Calculated
Saad Khawar – Case Study - I - QMDM 6

Min E ( Objective) 0.905


1 wg + wu + wc+ ws = 1 1 1.000 0.000
2 48.14wg + 34.62wu + 36.72wc + 33.16ws ≥ 36.72 36.72 36.720 0.000
3 43.10wg + 27.11wu + 45.98wc + 56.46ws ≥ 45.98 45.98 45.980 0.000
4 253wg + 148wu + 175wc + 160ws ≥ 175 175 176.615 1.615
5 41wg + 27wu + 23wc + 84ws ≥ 23 23 60.027 37.027
6 285.20wg+ 162.30wu+ 275.70wc+ 210.40ws-275.70E ≤ 0 0.00 -35.824 35.824
7 123.80wg+ 128.70wu+ 348.50wc+ 154.10ws-348.50E ≤ 0 0.00 -174.422 174.422
8 106.72wg+ 64.21wu+ 104.10wc+ 104.04ws-104.10E ≤ 0 0.00 0.000 0.000
9 ws>0 0.00 0.527 0.000
10 wg>0 0.00 0.212 0.000
11 wc>0 0.00 0.000 0.000
12 wu>0 0.00 0.260 0.000
13 E>0 0.00 0.905 0.000

Proof
wg 0.212
wu 0.260
ws 0.527
wc 0.000
E 0.905

Putting in the inequalities


253wg + 148wu + 175wc + 160ws ≥ 175
253(0.212) + 148(0.260) + 175(0) + 160(0.527) ≥ 175
176 ≥ 175
Hence proved

Now checking for E


106.72wg+ 64.21wu+ 104.10wc+ 104.04ws-104.10E ≤ 0
106.72(0.212) + 64.21(0.260) + 104.10(0) + 104.04(0.527) - 104.10(0.905) ≤ 0
-0.06 ≤ 0
Hence proved
Saad Khawar – Case Study - I - QMDM 7

4. The conclusion
General hospital wg=0.212, University Hospital wu=0.260, State hospital ws=0.527 and County hospital
wc=0.000. Each input and output of the composite hospital is determined by the same weighted average
of the inputs and outputs of these three hospitals. The slack/surplus column provides some additional
information about the efficiency of County hospital compared to the Composite hospital. Specifically,
the Composite hospital has at least as much of each output as County hospital has (constraints 2-5) and
provides 1.6 more nurses trained (surplus for constraint 4) and 37 more interns trained (surplus for
constraint 5). The slack of 0 from constraint 8 shows that the Composite hospital uses approximately
90.5% of the bed-days used by County hospital. The slack values for constraints 6 & 7 shows that less
than 90.5% of the FTE non physician and the supplies expense resources used at County hospital are
used by Composite hospital.

Clearly, the Composite hospital is more efficient then County hospital, and we are justified in concluding
that County hospital is relatively inefficient compared to the other hospitals in the group. Given the
results of the DEA analysis hospital administrators should examine operations to determine how County
hospital resources can be more effectively utilized.

You might also like