Anticipating the demand for supermarket products
Managing a supermarket’s stock is not an easy task. It is important to be able to predict the demand for products as much as possible to ensure optimal supply from the purchasing center.
So, how can we anticipate demand to correctly define our safety and alert stocks?
The seasonal coefficients method
This very versatile method will allow you, depending on the precision with which you apply it, to detect the trends in demand for products on a daily, weekly, and by extension, seasonal basis.
The method consists of 4 steps:
- Define the “season”
Do you want to analyze the demand by day of the week? By month of the year? In other words, for example, you want to try to predict the general level of demand for Monday, Tuesday, Wednesday, Thursday, Saturday, Sunday.
- Calculate the Average Demand per Unit
If you choose the days of the week as a unit, you must take the maximum number of days for which you have demand data, sum them up, and then divide by the number of days. This will give you the average demand per day.
- Calculate the Average Demand per Season
You now need to calculate the average for each season. If you want to differentiate weekdays, you will need to calculate the average for all Mondays, Tuesdays, Wednesdays, etc.
- Calculate the Seasonal Coefficients
You will now divide the average for each day of the week by the average per day. This will give you what are called “seasonal coefficients”. For example, if you get 1.1 for Monday and 2.3 for Friday, you will know that there is a greater demand on Fridays than on Mondays.
For example, if you get 1.1 for Monday and 2.3 for Friday, you will know that there is a greater demand on Fridays than on Mondays.
If we start with an average quantity of Belle France products sold per day of 50, on Monday we will predict a stock of at least 50 x 1.1 = 55 and on Friday of 115. If the coefficient is 0.7 for Wednesday, the necessary stock is 35. Applied to perishable stock, this type of analysis can be very powerful.
The calculation can also apply to the months of the year and combine with the coefficients per day of the week. We will therefore have daily and monthly coefficients.
For example, imagine that for certain Belle France chocolate products the monthly coefficient for December is 3 and the daily (Monday) is 2 and the average daily demand is 10.
For a Monday in December, we will therefore have the predicted stock taking into account the day of the week of 2 x 10 = 20, to which we add the effect of the month of the year 20 x 3 = 60.
