Hi Ujwal,
This is too broad a question to ask but I will try explaining. it.
Data in demand planning is usually past sales orders, invoiced orders, open orders, backlogs etc. The main input is only past sales orders. The rest of the orders are informational purpose.
using the past sales orders over a period, the system tries to forecast how the future orders could be.
So,the input for DP is past sales orders and output is forecast.
Since not all products behave the same, we have to classify the products based on their growth trend andlife cycle. Some products could be growing continuously some my be mature, some declining...so you want to classify them as per the business input and forecast them individually using a different forecast profile. So, that is the reason why you don't have only on model doing all the forecast in a real world scenario. You have multiple models forecasting for different products and you may have some more models overwriting previous runs. This is totally dependent on business process.
Once you have the forecast, it is evaluated by the planner and made changes. After his changes, different departments, like finance, sales, marketing etc., get together to make adjustments. Once this is ready, it is released to either R/3 or SNP depending on the scenario.
Hope this helps.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.