## 1. Quantitative Methods

### 1.1. Quantitative forecasting models - Use mathematical techniques - Based on historical data - Can include causal variables - Less accurate as the forecast’s time horizon increases

1.1.1. Time Series Forecasting Models Assumption: The future is an extension of the past Historical data can be used to predict future demand.

1.1.1.1. Naïve Forecast

1.1.1.1.1. the estimate for the next period is equal to the actual demand for the immediate past period.

1.1.1.1.2. Ft+1 = At

1.1.1.1.3. Where

1.1.1.1.4. Pros

1.1.1.1.5. Cons

1.1.1.2. Simple Moving Average Forecast

1.1.1.2.1. uses historical data to calculate a moving average and works well when the demand is fairly stable over time.

1.1.1.2.2. Actual demand for period t +1 equal to Sum of actual demand from t to t-n+1 devided by n( number of periods used to calculate moving average) t>n

1.1.1.2.3. Pros

1.1.1.2.4. Cons

1.1.1.3. Weighted Moving Average Forecast

1.1.1.3.1. An n-period weighted moving average forecast is the weighted moving average of the n-period observations, using unequal weights.

1.1.1.3.2. Formular

1.1.1.3.3. Pros & cons

1.1.1.4. Exponential Smoothing Forecast

1.1.1.4.1. the forecast for the next period’s demand is the current period’s forecast adjusted by a fraction of the difference between the current period’s actual demand and forecast.

1.1.1.4.2. Ft+1= Ft + a(AF) = aA + (1-a)Ft)

1.1.1.4.3. When: Ft=forecast for period t A = actual demand for period t a = smoothing constant (0 ≤ a ≤1)

1.1.1.4.4. Pros

1.1.1.4.5. Cons

1.1.1.5. Linear Trend Forecast

1.1.1.5.1. using simple linear regression (trend line) to fit a line to a series of data occurring over time.

1.1.1.5.2. Formula: The trend line equation Ŷ = bo + b₁x

1.1.1.5.3. Ŷ= forecast or dependent variable x = time variable, also independent variable values b₁ = n Σ(xy)-Σx Σy, by = Σy-b₁ Ex y dependent variable values n = No. of observations

1.1.1.5.4. Pros

1.1.1.5.5. Cons

## 2. Forecast Accuracy

### 2.1. Forecast Error

2.1.1. is the different between the actual quantity and the forcast

2.1.2. et = At - Ft

2.1.2.1. Cause-and-Effect Forecasting Models Assumption: One or more factors (independent variables) are related to demand Can be used to predict future demand.

2.1.2.1.1. Simple Linear Regression Forecast

2.1.2.1.2. Multiple Regression Forecast

2.1.3. Where

2.1.3.1. At: actual demand for period t

2.1.3.1.1. et: forcast error of period t

2.1.3.2. Ft: forcast for period t

### 2.2. Measuare

2.2.1. MAD( mean absoltedeviation)

2.2.2. MAPE( Mean absolute percentage error)

2.2.3. MSE( Mean square error)

## 3. Qualitative Methods

### 3.1. Qualitative Forcasting Models

3.1.1. Jury of Executive Opinion

3.1.1.1. A meeting of SME to forcast the market

3.1.1.2. Apply for long range planning & new product introductions, general demand forcasting

3.1.1.3. Pros: Knowledgable and exprienced => Forecast valuable

3.1.1.4. Cons: value & realibity of outcome can be disminished

3.1.2. Delphi Method

3.1.2.1. From Round 1(expert respond) => Round 2(expert respond) => Round n( expert respond)=> Final round( reac consensus)

3.1.2.2. Apply for: H-Risk technology forcasting, large, expensive project, major new product introduction

3.1.2.3. Pros:Group members do not physically meet avoid the scenario where one or a few experts could dominate a discussion

3.1.2.4. Cons: time- consuming and very expensive

3.1.3. Sales Force Composite

3.1.3.1. Based on the knowledge of sales team about the market and estimates of customer needs.

3.1.3.2. Apply for: all kind of projects

3.1.3.3. Pros: The forecast tends to be reliable because salespeople are close to customers.

3.1.3.4. Cons: Individual biases could negatively impact the effectiveness of this approach.

3.1.4. Customer Surveys

3.1.4.1. 1. Design a forecasting questionnaire.

3.1.4.2. 2. Choose the target population.

3.1.4.3. 3. Carry out the survey through telephone, mail, Internet, or personal interviews.

3.1.4.4. 4. Collect and analyze data.

3.1.4.5. 5. Make forecasts from the results.