Sri Krishna College of Technology

(An Autonomous Institution)

Affiliated to Anna University | Approved by AICTE | Accredited by NAAC with ‘A’ Grade
Kovaipudur, Coimbatore, Tamil Nadu – 641 042.

NAME :LOGVARSHINI.A.R
REGISTER NO :L727824TPMB054
ASSIGNMENT NO :2

DATE OF
SUBMISSION : 23.10.2024

SUBJECT NAME : MANAGERIAL

ECONOMICS
NAME: LOGAVARSHINI A.R
REGISTER NUMBER: 727824TPMB054
ASSIGNMENT NUMBER: 2
DATE : 23.10.2024
TITLE : METHODS OF DEMAND
FORECASTING AND IT’S
EFFECTIVENESS

CONTENTS : 1. INTODUCTION
-About Demand
Forecasting in clothing
Industry
2. METHODS OF
FORECASTING
-Qualitative
-Quantitative
3. EVALUATION
4. CONCLUSION
 DEMAND FORECASTING IN CLOTHING INDUSTRY

INTRODUCTION

The clothing industry in India is dynamic and highly influenced by factors such as
seasonality, fashion trends, festivals, and consumer preferences. Accurate demand forecasting
is essential for clothing retailers to manage inventory, reduce costs, and meet consumer
demand effectively. Various qualitative and quantitative forecasting methods can be used to
predict future demand, each with its own advantages and limitations.

Now, we will explore these various forecasting methods, focusing on the Trend Projection
method from the quantitative techniques, and Consumers' Opinion Survey from the
qualitative techniques. We will also evaluate the effectiveness of these methods using
relevant evaluation tools to understand how well they work in the context of the Indian
clothing industry.

By examining these approaches, we will understand how retailers can optimize demand
forecasting to align with market dynamics, manage inventory efficiently, and ensure a
balance between supply and demand.
 METHODS OF FORECASTING

Qualitative Methods of Demand Forecasting

1. Consumers' Opinion Survey

In the Indian clothing industry, retailers use surveys and feedback from customers to
understand their preferences, especially for upcoming fashion trends or festive seasons. The
data gathered helps retailers forecast demand for specific types of clothing, such as ethnic wear
for festivals like Diwali or western fashion for youth.

This method is especially valuable for predicting demand for seasonal collections but can be
limited by consumers' inability to accurately predict their future purchases.

2. Sales Force Composite Method

Sales staff in the clothing industry are closely connected to customers' buying habits. They
can provide forecasts based on their observations of customer preferences, popular designs,
and sales trends during interactions in stores or online.

This method works well in retail clothing chains, where salespeople have a strong

understanding of what customers are likely to buy based on trends and recent promotions.
3. Experts’ Opinion Method
In this method, industry experts such as fashion consultants or market analysts provide
demand predictions based on broader market trends, economic conditions, and changes in
consumer behaviour

Particularly useful for high-end fashion brands or designers in India who must keep pace
with global fashion trends, while also catering to local tastes and preferences.

4. Group Discussion
Retailers bring together internal teams, such as design, marketing, and sales, to
collaboratively forecast demand. This discussion allows for diverse input, which is crucial
when forecasting for different segments, such as ethnic wear, casual wear, and formal wear.

Ideal for larger fashion brands or chains that need input from multiple departments, such as
marketing for understanding trends, and design for predicting which styles will be popular.

5. Delphi Technique
A structured process where retailers in the clothing industry seek opinions from a panel of
experts. After several rounds of anonymous surveys, a consensus is reached regarding future
demand trends.

Effective for predicting demand for new fashion lines or collections where the market
response is uncertain, such as when introducing eco-friendly clothing.
6. Market Simulation

This method involves simulating real market conditions by introducing new designs or
collections in a controlled environment. The response is then used to estimate future demand.

Suitable for testing consumer reactions to new fashion collections before a full-scale
launch, particularly in urban areas where fashion preferences shift rapidly.

7. Test Marketing
Clothing retailers introduce a limited collection in select markets to test its performance.
Feedback from these test markets helps refine the product and forecast broader demand.

Commonly used by Indian clothing retailers to test products in key cities such as Mumbai,
Delhi, and Bangalore before rolling out nationwide.

QUANTITATIVE METHODS OF DEMAND FORECASTING

1. Trend Projection
Historical sales data is analysed to identify long-term (secular), seasonal, cyclical, and
random trends. Clothing retailers use this method to understand how demand for certain types
of clothing, like winter wear or festival-specific outfits, fluctuates.

Useful for predicting demand during predictable cycles such as the festival season, summer
vacations, or winter months.
2. Smoothing Techniques
Smoothing techniques, such as moving averages, are used to minimize random fluctuations
in demand data. Clothing retailers use this method to smooth out irregularities in sales data,
providing a clearer picture of underlying demand patterns.

Effective for general clothing items where demand fluctuates less drastically, such as
everyday wear.

3. Moving Average
This method calculates the average demand over a specific number of past periods. In the
clothing industry, this is helpful for forecasting stable demand for basic clothing items like T-
shirts, jeans, or school uniforms.

Retailers can use moving averages to manage the inventory of staple items that have
relatively stable demand throughout the year.

4. Weighted Moving Average

In the weighted moving average method, more recent sales data is given higher
importance. This is useful for predicting demand in the fast-paced clothing industry, where
fashion trends can change quickly.

Particularly useful for fast-fashion brands that need to react quickly to shifting consumer
preferences.
5. Exponential Smoothing

Exponential smoothing assigns more weight to recent observations but includes all past
data in the forecast. In the clothing industry, it is used to predict short-term demand for fast-
moving fashion items.

Highly effective for brands dealing in fast-moving trends and seasonal products like
summer or winter collections.

6. Barometric Techniques
This method uses economic indicators such as consumer spending, GDP, and inflation
rates to predict future demand. Clothing retailers can use these indicators to understand how
economic conditions will affect purchasing power and demand.

Useful for forecasting during periods of economic uncertainty or during global events like
recessions or pandemics.

7. Econometric Methods
Econometric models use statistical techniques to forecast demand based on a range of
variables, including price, income levels, and competitor behaviour.

Effective for large retailers who want to predict how changes in the economy will affect
demand for various clothing lines.

8. Regression Analysis
Regression analysis establishes relationships between dependent and independent variables.
In the clothing industry, retailers might use regression to predict how changes in price or
marketing efforts influence sales. Useful for analysing how factors like promotions,
advertising campaigns, or seasonal discounts impact clothing sales.
 Common Evaluation Methods for All Forecasting Techniques

1. Mean Absolute Deviation (MAD)

Measures the average magnitude of errors in demand forecasts. A lower MAD

indicates more accurate forecasts.

2. Mean Squared Error (MSE)

Gives more weight to larger errors by squaring the forecast errors, helping
retailers identify large deviations.

3. Tracking Signal

Monitors forecast bias by comparing cumulative errors to the MAD. A

consistent tracking signal indicates the reliability of the forecasting method.

4. Mean Absolute Percentage Error (MAPE)

Expresses the accuracy of demand forecasts as a percentage, making it easier

to interpret forecast accuracy across different product categories.
 Evaluation of Demand Forecasting in the Indian Clothing Industry

Quantitative Method:

Exponential Smoothing

Exponential smoothing is highly effective in predicting short-term demand for fashion items,
particularly in the fast-moving clothing industry where trends shift quickly. This method
adjusts forecasts based on the most recent data, making it well-suited for the dynamic nature
of fashion.

Evaluation Process:

To evaluate the effectiveness of exponential smoothing in the clothing industry, we calculate

the MAPE by comparing forecasted demand for a set of clothing items (e.g., T-shirts, summer
wear) over a given period with the actual sales data.

For example, if a retailer predicted demand for summer wear using exponential smoothing,
and the actual sales data became available after the season, the forecast errors could be
calculated. The MAPE would provide a percentage measure of how accurate the forecasts
were.

Result:

Let’s assume the forecast for a particular clothing line resulted in a MAPE of 8%. This
indicates that on average, the forecast deviated by 8% from the actual demand. A lower
MAPE suggests that exponential smoothing has been fairly accurate in predicting demand for
these fashion items, but further adjustments may still be necessary for refining forecasts.
Qualitative Method:

Consumers' Opinion Survey

Consumers’ opinion surveys are a common qualitative method used to understand customer
preferences and predict demand for fashion products in India, especially during festive
seasons or when launching new collections.

Evaluation Process:

A cost-benefit analysis evaluates whether the cost of conducting consumer surveys justifies
the improvement in demand forecasting. In this case, the cost of designing and distributing
surveys (both online and offline) and analysing the data is compared against the benefits,
such as improved inventory planning, reduced stockouts, and higher customer satisfaction.

For example, if a clothing retailer conducted a survey before Diwali to gauge demand for
ethnic wear, the cost of the survey might include design, distribution, and analysis, totaling
₹50,000. If the improved demand forecast prevented excess inventory worth ₹300,000, the
benefits clearly outweigh the costs.

Result:

The retailer could determine that the ₹50,000 spent on the survey provided a net benefit of
₹250,000 by avoiding overstocking and markdowns. This demonstrates that consumers’
opinion surveys can be a valuable tool when used for targeted products, such as festive
collections in the Indian clothing industry.
Conclusion

In the Indian clothing industry, both qualitative and quantitative methods of demand
forecasting play a crucial role. While qualitative methods are more suitable for predicting
demand for new or seasonal fashion lines, quantitative techniques provide accurate
predictions for stable demand patterns. Combining these methods, along with continuous
evaluation, allows retailers to minimize errors, optimize inventory, and enhance overall
business performance.

REFERENCES:

1. Makridakis, S., Wheelwright, S. C., & Hyndman, R. J. (1998). Forecasting Methods and
Applications (3rd ed.). John Wiley & Sons.

2. Armstrong, J. S. (2001). Principles of Forecasting: A Handbook for Researchers and

Practitioners. Springer Science & Business Media.

Two Website Links

1. https://www.investopedia.com/terms/d/demand-forecasting.asp

2. https://corporatefinanceinstitute.com/resources/knowledge/modeling/demand-forecasting-
methods/