Operation Management

MT OM BBA501
BY: Dr. Anupam Jain
Associate Professor-ABS
Amity University, Jaipur.
 Module II
Demand Forecasting & Capacity
Planning
BY: Dr. Anupam Jain
Associate Professor-ABS
Amity University, Jaipur.
 Meaning of Demand Forecasting

Demand Forecasting is the process of estimating the future

demand for a product or service over a specific period, based on
historical data, market trends, and other influencing factors.
It is essential for production planning, inventory management,
workforce scheduling, and budgeting.
Importance of Demand Forecasting

Aligns production with market demand.

Avoids underproduction (stockouts) and overproduction

(excess inventory).

Helps in capacity planning and resource allocation.

Supports strategic decision-making.

 Types of Demand Forecasting

1 QUALITATIVE 2 QUANTITATIVE
FORECASTING METHODS FORECASTING METHODS
 Types
QUALITATIVE FORECASTING QUANTITATIVE
METHODS: FORECASTING METHODS:
1. Educated Gues 1. Linear Regression
2. Consensus Method 2. Moving Average
3. Delphi Method 3. Exponential Smoothing
4. Historical Analogy
5. Market Research
 Educated Guess
•Based on the experience and intuition of a manager or expert.
•Quick and inexpensive.
•Suitable for short-term forecasting in stable environments.
•Example: A store manager estimating festive season sales based on past years’ patterns.
Advantages:
➢ Quick
➢ low-cost
➢ Simple
Disadvantages:
➢ High risk of bias
➢ lacks scientific basis
 Consensus Method
• Group of experts collectively agree on a forecast.
• Involves open discussion until a common decision is reached.
• Often used in new product launches.
• Example: Senior managers in a company agreeing on quarterly sales targets.
Advantages: Combines diverse viewpoints
➢ encourages agreement.
Disadvantages:
➢ Risk of “groupthink,”
➢ dominant voices may overshadow others.
 Delphi Method
• Structured forecasting technique to avoid bias.
• Experts answer questionnaires anonymously in multiple rounds.
• Responses are aggregated and shared after each round.
• The process continues until consensus is reached.
Advantages:
➢ Eliminates influence of dominant individual
➢ Systematic approach
Disadvantages:
➢ Time-consuming
➢ Requires commitment from experts.
 Historical Analogy
• Uses the demand pattern of a similar product or service as a reference.
• Example: Forecasting demand for electric bikes using historical sales data of electric
scooters.
Advantages:
➢ Useful for new products with no direct history.
Disadvantages:
➢ Assumes similarity between products
➢ May ignore unique factors.
 Market Research
• Collecting primary and secondary data about consumer needs, preferences, and buying patterns.
• Methods: surveys, interviews, focus groups, observation.
• Example: FMCG companies surveying customers before launching a new snack.
Advantages:
➢ Provides direct consumer insights
➢ Reliable for short-term predictions.
Disadvantages:
➢ Expensive
➢ Time-consuming.
 Linear Regression
•Linear Regression is a statistical method used to establish a relationship
between two or more variables.
•In demand forecasting, it helps to predict future demand (Dependent
Variable) based on time or other factors (Independent Variable).
•The assumption is that the relationship between demand and influencing
factors is linear.
• We can say that linear regression is used to find out the most probable
value of X on a given value of Y. Here, the value of Y is independent, and X is
dependent.
•Example: Predicting sales based on advertising expenditure.
 Formula of Linear Regression
Y=a+bX Steps in Linear Regression Forecasting
1. Collect Data – Gather historical demand data and
Where: corresponding independent variable(s).
2. Plot Data – Create a scatter diagram to visualize
Y = Dependent variable (e.g., Demand) relationship.

X = Independent variable (e.g., Time,

Price, Income)
a = Intercept (value of Y when X = 0)
b = Slope (change in Y due to one unit
change in X) 5. Formulate Regression Equation – Substitute values of
a and b.
6. Forecast Demand – Insert the required X value into the
equation.
 Example: Simple Linear Regression

Given monthly sales data (X = month, Y = demand):

Regression Equation: Y = a + bX

Suppose: Y = 50 + 5X
• For Month 6, Forecast = 50 + 5(6) = 80 units
• For Month 10, Forecast = 50 + 5(10) = 100 units
 Example 1
Find the year-wise Historical sales data (units) as below:
Year (X) Demand (Y)
1 120
2 130
3 150
4 170
5 180

Find the value of year 6.

Solution 1

Note: Submission
XY=2410 and final
answer should be 198
 Example 2
Suppose demand (Y) depends on time (X in months).as below:
Months (X) Demand (Y)
1 10
2 12
3 13
4 16
5 18

Find the value of month 6.

Solution 2

Note: Submission
XY=227 and final
answer should be 19.8
 Advantages & Disadvantages
Advantages: Limitations:
• Simple to apply and interpret. • Assumes a linear relationship, may
not always hold.
• Shows both trend and causal
relationships. • Sensitive to outliers.
• Useful for short and medium-term • Requires sufficient historical data.
forecasting.
• Not suitable for long-term forecasting
• Provides objective and quantitative
forecasts.
 Moving Average Method

The Moving Average Method is a quantitative time-series forecasting

technique that smooths short-term fluctuations and highlights long-term
trends.
•It calculates the average of a fixed number of past data points (periods).
•As new data becomes available, the oldest observation is dropped, and the
newest is added — hence the “moving” nature.
Formula
 Steps to Calculate

Calculate the Drop the oldest

Choose the period Plot forecasts to
Collect past average of the data point as new
(e.g., 3-month, 4- compare with
demand data. chosen number of data arrives and
month, etc.). actual demand
past data points. recalculate.
 Example

Suppose the demand data for 6 months is:

10, 12, 11, 13, 15, 16
Using a 3-month moving average:
•Forecast for Month 4 = (10 + 12 + 11) ÷ 3 = 11
•Forecast for Month 5 = (12 + 11 + 13) ÷ 3 = 12
•Forecast for Month 6 = (11 + 13 + 15) ÷ 3 = 13
•Forecast for Month 7 = (13 + 15 + 16) ÷ 3 = 14.67
Practical Applications

Inventory Production
management. planning.

Short-term
Sales
demand
forecasting.
estimation.
 Advantages & Disadvantages
Advantages: Limitations:
• Easy to understand and apply. • Requires storing historical data.
• Useful for stable demand without • Less effective when demand has a
strong trend or seasonality. trend or seasonal pattern.
• Smooths out random fluctuations. • Lags behind actual changes (since
it’s an average of past values).
 Exponential Smoothing Method
•Exponential Smoothing is a time series forecasting technique that
applies decreasing weights to past observations.
•Unlike simple moving averages, it gives more importance to recent
data while still considering older data with gradually declining weights.
•It is widely used in business, production, and inventory management
due to its simplicity and efficiency.
 Concept

•Forecasting using exponential smoothing is based on the

idea that the most recent demand carries more relevance
in predicting the future.
•The weights applied to past data decrease exponentially,
never reaching zero but becoming negligible for very old
data.
Formula
 Explanation of Smoothing Constant (α)
•α is the weight assigned to the most recent demand.
•High α (close to 1):
➢ Forecast reacts quickly to recent changes but may fluctuate more.
➢ Suitable for environments with high variability.
•Low α (close to 0):
➢ Forecast reacts slowly, smoother predictions, but lags behind actual changes.
➢ Suitable for stable demand patterns.
•The value of α is chosen based on the stability and variability of data
(commonly between 0.1 and 0.3).
Example 1
Example 2
 Applications
•Inventory Management – To estimate future product
requirements.
•Production Planning – To align capacity with forecasted
demand.
•Retail Sales Forecasting – For short-term product demand.
•Service Operations – For workforce scheduling (e.g., call
centers, healthcare).
 Advantages & Disadvantages
Advantages: Limitations:
•Simple and Easy to Use – Requires little •Lag in Forecasting – It may lag behind
data (just last demand and forecast). trends if demand changes rapidly.
•Low Data Storage – Unlike moving •Single Variable Dependency – Only
averages, no need to store large historical considers past demand, ignoring other
datasets. influencing factors (e.g., promotions,
•Recent Data Focus – Quickly adapts to seasonality).
new changes in demand. •Choice of α is Crucial – Poor selection can
•Flexible – α can be adjusted to control lead to inaccurate forecasts.
sensitivity. •Not Suitable for Strong Seasonality or
Cyclic Data – Needs modified versions
(Holt’s or Winter’s methods).
Demand Forecasting & Capacity Planning – Quick Reference Flowchart

Step 1: Identify Forecasting Need

Step 2: Select Method (Qualitative /

Quantitative)

Step 3: Collect Data

Step 4: Apply Forecasting Model

Step 5: Analyze & Interpret Results

Step 6: Plan Capacity (Lead, Lag,

Match)

Step 7: Monitor & Revise

 Production Capacity

Production capacity refers to the maximum output that a

business, plant, or production system can produce in a given
period under normal operating conditions with the available
resources (machines, labor, materials, and technology).
It is a measure of the potential ability of a production system to
meet demand.
 Key Aspects of Production Capacity
Theoretical Capacity
•The maximum output a facility can produce if it operates continuously at full efficiency, without
any downtime.
•Often unrealistic as it ignores practical limitations (maintenance, breakdowns, employee rest,
etc.).
Practical Capacity (Effective Capacity)
•The achievable output considering real-world limitations, such as maintenance schedules, setup
time, labor constraints, and holidays.
•More realistic measure for planning.
Actual Capacity (Achieved Output)
•The real output delivered by the system during a specific period.
•Often less than effective capacity due to unforeseen issues (machine breakdowns, absenteeism,
material shortages, etc.).
 Factors Affecting Production Capacity

Technolo Labor Manage

Plant Material Maintena Product
gy and Availabili ment
Size and Availabili nce
Layout –
Machiner ty and
ty Practices
Efficienc Mix.
y –. Skill. y.
 Importance of Production Capacity

•Helps in meeting customer demand efficiently.

•Guides decisions on expansion, investment, and resource
allocation.
•Provides a basis for capacity planning and scheduling.
•Ensures cost efficiency by balancing demand and available
resources.
•Affects competitiveness and profitability of the firm.
 Capacity Planning

Capacity Planning is the process of determining the production

capacity needed by an organization to meet changing demands
for its products or services. It ensures that resources like
manpower, machines, raw materials, and technology are
optimally utilized to meet customer requirements without
underutilization or overburdening.
 Objectives of Capacity Planning
•Meet Demand Efficiently → Ensure production capacity is aligned with
market demand.
•Optimize Resource Utilization → Avoid underuse (idle resources) or
overuse (bottlenecks, breakdowns).
•Cost Efficiency → Reduce cost per unit by maintaining the right level of
capacity.
•Flexibility → Respond quickly to demand fluctuations or market
changes.
•Strategic Growth → Support long-term expansion and competitiveness.
 Types of Capacity Planning
Long-Term Capacity Planning
• Time horizon: More than 2 years.
• Focus: Expansion of facilities, addition of new plants, technology upgrades.
• Example: Setting up a new factory to meet demand growth.
Medium-Term Capacity Planning
• Time horizon: 6 months to 2 years.
• Focus: Workforce planning, subcontracting, equipment purchase, shift adjustments.
• Example: Hiring seasonal workers for upcoming demand.
Short-Term Capacity Planning
• Time horizon: Daily to 6 months.
• Focus: Adjusting workforce schedules, machine loading, overtime, inventory control.
• Example: Running an extra shift to complete urgent orders.
 Capacity Planning Strategies
Lead Strategy (Proactive)
•Add capacity in advance of expected demand.
•Advantage: No missed demand, competitive edge.
•Risk: Excess idle capacity if demand does not materialize.

Lag Strategy (Reactive)

•Add capacity only when demand is proven.
•Advantage: Less risk of overcapacity.
•Risk: May lose customers due to unfulfilled demand.

Match Strategy (Balanced)

•Gradual expansion in small increments as demand increases.
•Advantage: Reduces both risks of overcapacity and undercapacity.
•Widely used strategy.
 Steps in Capacity Planning
•Forecast Demand – Estimate future demand using qualitative or quantitative
forecasting methods.
•Assess Current Capacity – Measure the current output levels, machine utilization,
labor skills.
•Identify Gaps – Compare demand forecast with existing capacity to find
shortfall/excess.
•Develop Alternatives – Options include overtime, subcontracting, expansion, new
equipment.
•Evaluate Alternatives – Check feasibility, cost, flexibility, and risk.
•Select Best Option – Choose the most cost-effective and flexible capacity plan.
•Implement & Monitor – Execute the plan and continuously monitor performance.
 Challenges in Capacity Planning

•Demand uncertainty (seasonal fluctuations, changing customer

preferences).
•High cost of adding/removing capacity.
•Long lead time for equipment and facility expansion.
•Balancing short-term efficiency with long-term flexibility.
Production Capacity Planning – Quick Reference Flowchart

Step 1: Forecast Demand

Step 2: Determine Required Capacity

Step 3: Assess Existing Capacity

Step 4: Identify Capacity Gap

Step 5: Choose Strategy (Lead / Lag /

Match)

Step 6: Implement Capacity Changes

Step 7: Monitor Performance & Adjust