If it was difficult to predict demand trends fifty years ago, then there is not even a word for how challenging it is to predict demand in today’s dynamic market. But Rakesh Singh writes for Business Standard that an element of forecasting is necessary for supply chain success. Right now, most organizations are still doing a lousy job implementing demand uncertainty into production planning processes.
Demand to Know
Firm-level variables must be identified as the first step of good forecasting. For too long, subjective variables were used that were arguably no more clear than the things they were supposed to forecast. Which types of variables will be useful for you will depend on a case-by-case basis of course.
One big company found that its vacuum cleaners and spare part requirement were tied to somewhat predictable seasonality, though the company only realized this after a time series technique and an ERP system both failed. Agricultural businesses that sell tractors find a clear correlation between sales and the forecast for crops in each given region. There persists a problem though where when forecasting does not seem implicit from the data available, the business assumes forecasting is impossible rather than finding fault with their own business practices and analysis. Businesses should invest in software with a more collaborative architecture if they want to find data on pertinent variables faster.
For further elaboration, you can read the full article here: http://www.business-standard.com/article/management/demand-forecasting-in-a-supply-chain-114081700501_1.html