Question

B. Inventory management
The current policy is to order 100,000 units when the inventory level falls to 35,000 units. However, forecast demand to meet market requirements for next year is 625,000 units. The cost of placing and processing an order is R250, while the annual cost of holding a unit in stores is 5% of the unit purchase price. Both costs are expected to be constant during the next year. Shop n Pay sells a unit of the product for R15.00 at cost plus 50%. Orders are received two weeks after being placed with the supplier. You should assume a 350-day year and that demand is constant throughout the year.

C. Forecasting and other supply chain activities A recent report produced by Shop n Pay’s chief demand planner has revealed that the retailer’s reliance on simple exponential smoothing to forecast demand for its products is partly responsible for the significant bullwhip effect in the supply chain and the stock outs in the past few years. In this regard, the chief demand planner is recommending that exponential smoothing with trend adjustment (Holt’s Model) should be used to forecast demand for aggregate planning for the 2020 fiscal year.

Mr Msimanga has retrieved from the retailer’s database the following information on number of goods sold over a two–year period. Year Monthly sales (‘000 units)

Jan 2018 -650 /2019 -750

Feb 2018-700 /2019 -800

Mar 2018 -820/2019-900

Apr 2018 -850/2019-920

May 2018-700/2019- 680

Jun 2018-840/2019-950

Jul 2018-930/2019-1080

Aug 2018-630/2019-742

Sep 2018-860/2019-920

Oct 2018-600 /2019 -705

Nov 2018-1050/2019 -1100

Dec 2018 - 750 /2019- 820

2.3 REQUIRED: Use the information provided above under section C:

Forecasting and other supply chain activities to answer the following questions:

2.3.1 “A recent report produced by Shop n Pay’s chief demand planner has revealed that the retailer’s reliance on simple exponential smoothing to forecast demand for its products is partly responsible of the significant pile-up of goods in the company’s main distribution centre.”

2.3.2
Calculate the Exponential smoothing with trend component forecast (FIT) for second to the twenty-fifth month using an initial trend forecast (?1) of 10 000, an initial exponential smoothing forecast (?1) of 640 000 units, ?=0.20, and ?=0.30.
2.3.3
Plot the actual demand, the simple exponential smoothing (SES) forecast, and trend-adjusted exponential smoothing (FIT) forecast on the same graph. Briefly comment on the observed difference(s) between the plots of the two forecasts.

Answer 1

Answer:

a) Please refer to the following calculation snapshots:

Adjusted Forecast for any month = Trend for that month + Forecast for that month

Trend for any month = Trend for previous month + Beta * (Forecast for this month - Adjusted Forecast for previous month)

Forecast for any month = Adjusted Forecast for previous month + Alpha * (Actual sales of previous month - Adjusted Forecast for previous month)

.

For Simple Exponential Smoothing,

forecast for any month = forecast for previous month + alpha * (Actual sales of previous month - forecast for previous month)

b) From the graph, it is evident that the Simple Exponential Forecast (SEF) fails to capture the trend with the monthly sales, thereby leading to higher errors (difference in actual sales and forecast) for subsequent months.

Another evident thing is, before SEF does not include the trend component like the adjusted exponential forecast, SEF tends to average out the forecast and hence, in cases when sales are dynamically rising and falling in back to back months, the losses could be higher for adjusted exponential forecast as it tries to include the trend and get the adjusted forecast for next month i.e. the effect of trend for this month could be seen in the next month whereas SEF does not have high variability, thereby averaging out the forecast with respect to the variability in sales.

This can be seen for the month of May 2018 where the actual sales have dropped to 700 from 850 in April and SEF is 719.50 whereas Adjusted Forecast is 781.31.

Similarly for the month of Jan 2019 where the actual sales have remained constant to 750 and SEF is 789.54 whereas Adjusted Forecast is 862.37 precisely due to the accumulated trend from the past forecasts.

.

Do let me know in case of any queries. Happy Learning!