PROJECT RISK MANAGEMENT

🌟 1. Meaning of Project Risk Management

 Project Risk Management involves identifying, assessing, and
responding to uncertainties that may affect a project’s objectives — cost,
time, and performance.
 It helps managers evaluate how project outcomes vary when
assumptions change.
 Techniques include:
o Simple Sensitivity Analysis

o Scenario Analysis

o Monte Carlo Simulation

o Decision Tree Analysis

🔹 2. SIMPLE SENSITIVITY ANALYSIS

🧠 Concept
Sensitivity analysis studies how a change in one key variable (e.g., sales,
cost, discount rate) affects project measures such as NPV, IRR, or Payback
while keeping other variables constant.
It answers:
“Which factor has the biggest impact on NPV or project outcome?”

🧾 Steps
1. Identify base case (expected) values of inputs — e.g., sales, cost, rate.
2. Compute base-case NPV.
3. Change one variable at a time by a fixed % (say ±10%, ±20%).
4. Recalculate NPV for each change.
5. Compare how much NPV changes.

💡 Numerical Example
 Base
Item
Value

Initial
₹1,00,000
Investment

Annual Cash
₹30,000
Inflow

Project Life 5 years

Discount Rate 10%

Base-case NPV:
NPV = (PV of inflows) − Outflow
PVIFA(10%,5) = 3.791
NPV = 30,000 × 3.791 − 1,00,000 = ₹13,730
Now, vary one factor — cash inflow ±20%

Cash Chang
NPV (₹)
Inflow e

24,000 (24,000 × 3.791) − 1,00,000 = ↓147

(−20%) −₹9,016 %

30,000
₹13,730 —
(Base)

36,000 (36,000 × 3.791) − 1,00,000 = ↑166

(+20%) ₹36,476 %

Interpretation:
The NPV is very sensitive to cash inflows, so sales volume is a critical risk
factor.

✅ Key Takeaways
 Simple, easy, but examines one variable at a time.
 Does not show combined effect of simultaneous changes.
 Useful for first-level risk identification.

🔹 3. MONTE CARLO SIMULATION

🧠 Concept
Monte Carlo Simulation (MCS) is a probabilistic risk analysis technique.
Instead of using one value per variable, we assign probability distributions
(ranges) to uncertain variables (like demand, cost, rate of return) and then
simulate thousands of random combinations.
It gives a probability distribution of NPV rather than a single value.

⚙️Steps (conceptually)
1. Identify uncertain variables (e.g., demand, cost, rate).
2. Specify probability distributions for each (normal, triangular, etc.).
3. Generate random combinations (using computer simulation).
4. Compute project NPV for each set.
5. Analyze distribution of NPV outcomes (mean, standard deviation,
probability of NPV < 0).

💡 Simplified Illustrative Example

A project has uncertain annual cash inflow depending on market conditions:

Probabili Cash Inflow

Case
ty (₹)

Pessimisti
0.2 25,000
c

Most
0.5 30,000
likely

Optimistic 0.3 40,000

Other data:
Investment ₹80,000, Project life = 4 years, Discount rate = 10%
We can simulate a few possible outcomes (for simplicity, 3 cases):

Scenari Cash PV Factor

NPV (₹)
o Inflow (10%,4)=3.17

25,000×3.17 − 80,000 = −
1 25,000 3.17
₹750

30,000×3.17 − 80,000 =
2 30,000 3.17
₹15,095

40,000×3.17 − 80,000 =
3 40,000 3.17
₹46,794

Expected NPV = (−750×0.2) + (15095×0.5) + (46,794×0.3)

= ₹21,435
So Expected NPV = ₹21,435
but the simulation (if repeated 1,000 times) would give the probability that NPV
is negative or positive.
✅ Key Takeaways
 Provides a range and probability of NPVs instead of one number.
 Useful when many variables are uncertain.
 Requires computer or Excel simulation tools (e.g., @RISK, Crystal Ball).
 Limitations: needs statistical input and software; harder for manual
calculation.

🔹 4. DECISION TREE ANALYSIS (DTA)

🧠 Concept
A Decision Tree is a diagram showing sequences of decisions and possible
outcomes, useful when future events depend on earlier choices.
It quantifies risk and expected values for alternative project strategies.

⚙️Steps
1. Define the decision alternatives (e.g., launch project, delay, abandon).
2. Define possible states of nature (good demand, poor demand, etc.) with
probabilities.
3. Estimate payoff (NPV or profit) for each combination.
4. Multiply payoff × probability to get Expected Monetary Value (EMV) for
each branch.
5. Choose the decision with highest EMV.

💡 Example (Decision Tree Numerical)

A company considers a new product project costing ₹10 lakh.
There are two demand scenarios if launched immediately:

Probabili NPV (₹
Scenario
ty lakh)

High
0.6 +₹8
Demand

Low
0.4 −₹4
Demand

Alternatively, the company can delay by 1 year and conduct a market survey
costing ₹1 lakh**,** which predicts market conditions more accurately:
If survey result is favourable (prob = 0.7) → proceed with project:
 High Demand (prob 0.8): NPV +₹8
  Low Demand (prob 0.2): NPV −₹4
If survey result is unfavourable (prob = 0.3) → cancel project (NPV = 0).

🔢 Solution (Step-by-Step)
Option 1: Launch now
EMV = (0.6 × 8) + (0.4 × −4) = 4.8 − 1.6 = ₹3.2 lakh
Option 2: Conduct survey first
For favourable survey:
EMV = (0.8 × 8) + (0.2 × −4) = 6.4 − 0.8 = 5.6 lakh
Net of survey cost: 5.6 − 1 = 4.6 lakh
Probability of favourable = 0.7 → expected = 0.7 × 4.6 = 3.22 lakh
Probability of unfavourable = 0.3 → NPV = 0 → 0.3 × 0 = 0
Total EMV (survey option) = 3.22 lakh

✅ Decision:

EMV (₹
Option
lakh)

Launch Now 3.2

Conduct
3.22
Survey

✅ Preferred Option: Conduct survey (slightly higher EMV).

✅ Key Takeaways
 Combines probabilities and decision logic.
 Visually shows sequential decisions and uncertainty.
 Helps justify “wait or invest now” choices.
 Best for discrete risk outcomes (e.g., high vs low demand).
 Limitation: gets complex with many branches.

🧩 5. Summary Comparison Table

Key
Technique Type Output Limitation
Advantage

Sensitivity Determinist Change in NPV Simple & visual Only one variable
Analysis ic with variable at a time
 Key
Technique Type Output Limitation
Advantage

change

Monte Carlo Probabilisti Distribution of Gives range & Requires software

Simulation c NPVs probability & statistical inputs

Decision Handles Becomes complex

Probabilisti Expected
Tree sequential with many
c Monetary Value
Analysis decisions branches

 📊 Quick Practice Exercise (for Students)

A project requires ₹2,00,000 investment. Cash inflows depend on market
conditions:

Scenari Probabili Annual Inflow

o ty (₹)

Boom 0.3 90,000

Normal 0.5 70,000

Recessio
0.2 50,000
n

Project life: 3 years, discount rate = 10%.

Compute Expected NPV using Monte Carlo approach (3 scenarios).
PVIFA(10%,3) = 2.486

Scenari
PV Inflows NPV Weighted NPV
o

90,000×2.486=2,23 +23,7
Boom 0.3×23,740=7,122
,740 40

70,000×2.486=1,74 −25,9 0.5×(−25,980)=−1

Normal
,020 80 2,990

Recessio 50,000×2.486=1,24 −75,7 0.2×(−75,700)=−1

n ,300 00 5,140

Sum Weighted NPV = 7,122 − 12,990 − 15,140 = −₹21,008

→ Expected NPV = −₹21,000 (unfavourable) → Reject project.

✅ Final Takeaway for Students

Concept Core Idea

Sensitivity
“What if one input changes?”
Analysis

Monte Carlo
“What’s the probability of different NPVs?”
Simulation

Decision Tree “Which decision path gives the highest expected

Analysis payoff?”