Demand Forecasting

Question 1

What is a forecast?

Answer 1

A prediction of what will occur in the future

Question 2

When are forecasts used for planning?

Answer 2

In situations where demand is uncertain (push situations)

Used to predict demand to schedule inventory, production, ordering etc.

Question 3

What is the Bullwhip Effect and how can it be related to demand forecasting?

Answer 3

A phenomenon where small fluctuations in customer demand cause increasingly larger fluctuations of orders upstream in the supply chain

• Ordering is amplified for each upstream supplier as they try to match the demand
• It is therefore important to have good demand forecasting strategies in order to prevent the bullwhip effect from occurring

Question 4

What are the 3 laws of forecasting?

Answer 4

1. Forecasts are always wrong
2. Detailed forecasts are worse than aggregate
3. The further into the future, the less reliable forecast

Question 5

What are the 5 basic patterns of demand in terms of time?

Answer 5

1. Horizontal: Data fluctuates around a constant mean/average
2. Trend: Systematic increase/decrease of average demand over time
3. Seasonal Regular, repeatable variations in demand
4. Cyclical: less predictable, long term fluctuations in demand
5. Irregular, unpredictable variations in data that cannot be forecasted

Question 6

How can you control or change demand patterns? Name 4 out of 6 options

Answer 6

1. Complementary products: even out the load on resources by producing products with different demand cycles
2. Promotional pricing: Campaigns used to increase sales –> increased demand
3. Prescheduled appointments: Encouraging customers to schedule their purchases or services, demand can be spread over time
4. Reservations: Similar to prescheduled appointment:
   - Manage demand and plan resources
5. Backlogs:
6. Backorders: Allow orders even if stockout occurs, fulfilled once stock is replenished

Question 7

What types of forecasting methods can be used?

Answer 7

Either

1. Qualititative/judgment methods or quantitative methods

Question 8

What are qualitative forecasting methods? Name 4 different techniques

Answer 8

• Often used when historical data on demand is unavailable or unreliable

1. Expert judgment: Experts estimates demand
2. Delphi Method: Several stakeholders collaborate iteratively to predict demand
3. Sales person estimate what sales will be in the region
4. Market Survey: Input from potential customers via market survey

Question 9

What are quantitative forecasting methods? Name 3 different techniques?

Answer 9

Methods that rely on historical data and mathematical models to predict demand

1. Time series forecasting: Statistical approach that relies on historical demand: recognize trends
2. Casual method: Collect historical data on independent variables like campaigns, competitors actions etc
3. Linear regression: combo os both methods: Identify trendlines by analyzing demand

Question 10

What is primary and secondary demand? And what is the difference?

Answer 10

Primary demand: The demand which comes from the market. (Independent demand)

Secondary demand: The demand for component and parts generated from the primary demand. (Dependent demand)

Question 11

What is independent demand?

Answer 11

Demand for products that are not influence by the demand of other products

Question 12

What is dependent demand?

Answer 12

Demand for products that are related to the production of finished goods/the demand for a finished product.

Tires for cars

Question 13

What are positive and negative errors when it comes to demand forecasting?

Answer 13

Postivite errors: Too low forecast
Negative forecast: Too high forecast

Question 14

What methods/indicators can be used to measure the accuracy of forecasts?

Answer 14

1. Mean Square Error (MSE)
2. Mean Absolute Deviation (MAD)
3. Mean absolute percent error (MAPE)

Question 15

What is direct and indirect demand?

Answer 15

Direct demand: The demand from end customer

Indirect demand: Demand by operators in the supply chain (not end customer involved)

Question 16

What is a by-product?

Answer 16

Secondary/Excess product from the manufacturing of another product that is not the main focus of the production

Question 17

Talking about Time Series analysis forecasting, what are different forecasting methods under that category of forecasting?

Answer 17

1. Naive method: The forecast= the previous demand
2. Moving average: Average of demand for a certain number of previous time periods
3. Weighted moving average: Gives more weight to more recent time periods. If 3: 0,5Dt+0,3Dt-1+0,2*Dt-2
4. Exponential smoothing: More recent data are still given more weight controlled by a smoothing factor alpha
5. Exponential smoothing with trend correction/Holts method:
   Exponential smoothing but also considers trends in data, two smoothing factors

Question 18

What is time series analysis forecasting?

Answer 18

Forecasting demand based on historical data. Future values can be estimated from past values

Question 19

Draw the different graphs for the five basic demand patterns

Answer 19

kolla facit boken