Data Envelopment Analysis (DEA)

1
Data Envelopment Analysis (DEA)
Identifying Efficient Decision Making Units
2
Measuring Service ProductivityBranch Bank Example
1. ACCOUNTING RATIO COST PER TELLER TRANSACTION
HIGHER RATIO WOULD BE INEFFICIENT
RELATIVE TO OTHERS, BUT COULD BE EXPLAINED
BY 1. MIX OF OUTPUTS (SELLING CDS VS
SIMPLE DEPOSITS) 2. MIX OF INPUTS (USE OF
ATMS AND LIVE TELLERS) 2. BROAD BASED MEASURE
RETURN ON INVESTMENT OVERALL IMPORTANT BUT
NOT SUFFICIENT TO EVALUATE OPERATING EFFICIENCY
OF INDIVIDUAL DECISION MAKING UNIT (DMU)
FOR EXAMPLE A PROFITABLE BRANCH
COULD BE THE RESULT OF HIGHER-THAN-AVERAGE
PROPORTION OF REVENUE GENERATING TRANSACTIONS
RATHER THAN COST-EFFICIENT USE OF
RESOURCES.
3
THE DEA MODELFractional Form

j 1,2,,M i 1,2,,N
4
DEA in Standard LP Form (Scaling inputs to sum
to 1.0)
5
Burger Palace Example
6
Burger Palace Productivity Frontier
S1 (2,200)
S2 (4,150)
S3 (4,100)
S4 (6,100)
S5 (8,80)
C (5.3, 88.9)
S6 (10,50)
7
Summary of DEA Results

8
DEA and Strategic Planning
Under-performing Benchmark
potential stars group
Problem Candidates
Branches for divestiture
High
Profit
Low
Low
High
Efficiency