How to Evaluate Airline Route Efficiency with DEA and Python

Case Study: Airline Route Efficiency

Data Preparation

In this case, 13 airlines (DMUs) are evaluated with three inputs (fleet size, fuel consumption, employee count) and two outputs (passenger‑miles and freight‑ton‑miles). The raw numbers for each airline are listed in the original source.

Programming Environment

<code>Python 3.7.4 + PuLP 2.0</code>

Execution Steps

Step 1 – Import Packages

<code>from pulp import LpProblem, LpMinimize, LpVariable, lpSum, value</code>

Step 2 – Build Data Sets

<code>K = ["A","B","C","D","E","F","G","H","I","J","K","L","M"]
I = ["Aircraft","Fuel","Employee"]
J = ["Passenger","Freight"]
# X and Y dictionaries contain the input and output values for each DMU</code>

Step 3 – Construct CRS DEA Model (input‑oriented dual)

<code>model = LpProblem('CRS_model', LpMinimize)
theta_r = LpVariable('theta_r')
lambda_k = LpVariable.dicts('lambda_k', lowBound=0, indexs=K)
model += theta_r  # dual formulation
# input constraints
for i in I:
    model += lpSum([lambda_k[k] * X[i][k] for k in K]) <= theta_r * float(X[i][K[0]])
# output constraints
for j in J:
    model += lpSum([lambda_k[k] * Y[j][k] for k in K]) >= float(Y[j][K[0]])</code>

Step 4 – Solve Model

<code>model.solve()</code>

Step 5 – Construct VRS DEA Model (input‑oriented dual)

<code>model = LpProblem('VRS_model', LpMinimize)
theta_r = LpVariable('theta_r')
lambda_k = LpVariable.dicts('lambda_k', lowBound=0, indexs=K)
model += theta_r
# same input and output constraints as CRS
# convex‑combination constraint for variable returns to scale
model += lpSum([lambda_k[k] for k in K]) == 1
model.solve()</code>

Step 6 – Extract Efficiencies

<code>OE_outputText = 'These are OE of all DMUs\n...'
TE_outputText = 'These are TE of all DMUs\n...'
SE_outputText = 'These are SE of all DMUs\n...'
for k in range(len(K)):
    OE_text, OE_val = getOverallEfficiency(k)
    TE_text, TE_val = getTechnicalEfficiency(k)
    OE_outputText += OE_text
    TE_outputText += TE_text
    SE_outputText += f'{K[k]}：{round(OE_val / TE_val, 3)}\n'
print(OE_outputText)
print(TE_outputText)
print(SE_outputText)</code>

Results show that airlines C, G, H, I, K and L achieve full efficiency (OE = TE = SE = 1), while D, J and M have lower overall and technical efficiencies, indicating that scale effects should be examined to improve their performance.

Conclusion

DEA is a reliable benchmarking technique for homogeneous decision‑making units such as airlines or bank branches. It provides relative efficiency scores but does not prescribe specific improvement actions, so additional analysis is needed to design concrete efficiency‑enhancement strategies.