Scribd
Upload
23 views2 pages

Feature Engineering

The document outlines key features for sales forecasting, including historical sales data, calendar features, promotions, external factors, customer segmentation, lead time, and supplier data. It emphasizes the importance of incorporating lagged sales data and external influences to improve forecasting accuracy and mitigate the bullwhip effect. Additionally, it highlights the role of customer behavior and collaboration among supply chain partners in aligning forecasts.

Uploaded by

pranilbanoth12
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
23 views2 pages

Feature Engineering

The document outlines key features for sales forecasting, including historical sales data, calendar features, promotions, external factors, customer segmentation, lead time, and supplier data. It emphasizes the importance of incorporating lagged sales data and external influences to improve forecasting accuracy and mitigate the bullwhip effect. Additionally, it highlights the role of customer behavior and collaboration among supply chain partners in aligning forecasts.

Uploaded by

pranilbanoth12
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

1 Historical Sales Data (Lagged Features)

 lag_1: Sales from the previous day.


 lag_7: Sales from the same day in the previous week.
 lag_30: Sales from the same day in the previous month.
2 Calendar Features
Include information about dates, holidays, weekends, and
special events.
 Holidays and events often cause spikes or dips in demand.
 Calendar features help the model anticipate these changes
and avoid overreacting to temporary fluctuations.
3 Promotions and Discounts
 Promotions can cause sudden spikes in demand, which may
trigger the bullwhip effect.
4. External Factors
Include external factors like weather, economic indicators, or
competitor actions
5. Customer Segmentation
Include features related to customer behavior or segmentation.
 customer_segment: Category indicating the customer
segment (e.g., loyal, occasional).
 purchase_frequency: Average number of purchases per
customer.
6. Lead Time and Order Data
Include lead times (time between placing and receiving an order)
and order history.
 Long lead times can amplify the bullwhip effect.
 Including lead time and order data helps the model account
for delays and adjust forecasts accordingly.
7. Supplier and Distributor Data
 Information asymmetry between supply chain partners can
amplify the bullwhip effect.
 Sharing supplier and distributor data helps align forecasts
across the supply chain.

You might also like