Demand Forecasting
DEMAND FORECASTING
Demand forecasting is the process of predicting future average sales on the basis of historical data samples and market intelligence. The volatility of demand from an average level is supplied from the safety inventory. Any forecast is likely to be wrong, so the focus should be on understanding the range of potential forecast errors and the level of safety inventory that will cater for peak demand. An important additional calculation is forecast bias. This is the cumulative sum of under- or over forecasting over a period of time. If, for example, the staff continually over-predict sales there will be a negative forecast bias, which will result in excess inventory and vice-versa. A good process for demand forecasting involves collecting information as far down the supply chain as possible. For example, orders or bookings are better than sales when planning inventories and production. When forecasting the average demand it must be adapted for products with seasonal patterns – for example, garden products in summer and festive products for Christmas. The customer demand needs to be anticipated and a final date for reordering determined. Historic patterns of sales volumes can be a valuable source of reference for identifying appropriate inventory and ordering dates.
In the retail sector, the weather can play a large part in consumers’ desire to go shopping and what they will buy. Care needs to be taken to manage inventory and potentially use sale or return as a way to reduce the risk of being left with inventory after the period of demand. The more often a supplier is unreliable, the greater is the tendency for its customers to hold more safety inventory to compensate for supplier incompetence. By changing suppliers or improving existing supplier reliability, a business can reduce its safety inventory levels in the knowledge that the supplier can support the business efficiently.
There are numerous forecasting packages available, all of which
Demand forecasting is the process of predicting future average sales on the basis of historical data samples and market intelligence. The volatility of demand from an average level is supplied from the safety inventory. Any forecast is likely to be wrong, so the focus should be on understanding the range of potential forecast errors and the level of safety inventory that will cater for peak demand. An important additional calculation is forecast bias. This is the cumulative sum of under- or over forecasting over a period of time. If, for example, the staff continually over-predict sales there will be a negative forecast bias, which will result in excess inventory and vice-versa. A good process for demand forecasting involves collecting information as far down the supply chain as possible. For example, orders or bookings are better than sales when planning inventories and production. When forecasting the average demand it must be adapted for products with seasonal patterns – for example, garden products in summer and festive products for Christmas. The customer demand needs to be anticipated and a final date for reordering determined. Historic patterns of sales volumes can be a valuable source of reference for identifying appropriate inventory and ordering dates.
In the retail sector, the weather can play a large part in consumers’ desire to go shopping and what they will buy. Care needs to be taken to manage inventory and potentially use sale or return as a way to reduce the risk of being left with inventory after the period of demand. The more often a supplier is unreliable, the greater is the tendency for its customers to hold more safety inventory to compensate for supplier incompetence. By changing suppliers or improving existing supplier reliability, a business can reduce its safety inventory levels in the knowledge that the supplier can support the business efficiently.
There are numerous forecasting packages available, all of which