You are on page 1of 2

DEMAND FORECASTING FOR PERISHABLE SHORT SHELF-LIFE HOMEMADE FOOD

AT iD FRESH FOOD

1. Problem Statement
Due to various factors, ID fresh food has a difficult time estimating demand at each location.
They used to have their own salesperson visit all the stores every day to keep track of the
SKUs on the shelf and the daily requirements. And the salesman did not utilise any precise
strategy in estimating demand; instead, he relied only on his instinct. One of the causes for
predicting inaccurately is these restrictions. Forecasting based on gut instinct was very
irrational, resulting in the return of surplus products to ID fresh food and the failure to service
the final 10% or 15% of stores in that beat. These were the key obstacles that ID fresh foods
had to overcome. Their ultimate objective was to streamline the process of estimating future
demand so that the distribution channel could function effectively.

2. Approach and techniques used for Demand planning & forecasting


Exponential smoothing is a univariate time series forecasting approach that may be extended
to data with a systematic trend or seasonal component. It is a very effective form of
predicting the future demand. The prediction is a weighted sum of past observations, but the
model explicitly uses an exponentially decreasing weight for past observations. There are
three main types of exponential smoothing techniques. A simple exponential smoothing,
double exponential smoothing and triple exponential smoothing. The method used here is the
simple exponential smoothing.
In the data given, three alpha values are being assumed for each beat and the average forecast
has been calculated. In which the alpha which shows the least variation when compared with
the demand will taken into considerations and the interpretations are as follows.

3. Summary of results
Sales beat 81 SKU 95- Demand vs Forecast
In case of beat no.81, we can observe that the forecasted demand shows significant dip twice
in a month, in the middle and towards the end of the month, respectively. We can infer that
the customers buying pattern in beat 81 fluctuates twice in a month. Hence, we must keep
less stock at these points of the month when the demand is comparatively low.
Sales beat 98 SKU 95- Demand vs Forecast
In case of beat no.98, The demand for our products is varying between two extremes that is
from 90 units to as low as 49 units. And this fluctuation in demand can be seen happening
multiple times across the month. We can infer that beat 98 is a volatile route.

4. Recommendation
DEMAND FORECASTING FOR PERISHABLE SHORT SHELF-LIFE HOMEMADE FOOD
AT iD FRESH FOOD

For beat no.81, we would recommend that the salesperson carry around 100 units during the
mid-three dates of the month i.e., 14,15,16 and one week before every month end. Keeping in
mind the high demand expected during the holiday season in the month of December, we
recommend the salesperson carry up to 200 units.
When it comes to beat no.98, the consumption seems to be on the lower end averaging
around 70 units per day. We recommend, the sales party to carry about 80 to 90 units
everyday to be on the safer side gradually increasing it to 120 by the end of the month.

Poornima Sai- 2020JULB01056


Dineshwar Annadurai- 2020JULB01135
Aldrin D’cruz- 2020JULB01165

You might also like