PROJECT SUBMISSION
Task 1
● Method of forecasting: I would select Time Series Forecasting Method Precisely 3
Period Weighted Moving Average Method to forecast the demand for the period of 6
months. Since, this method is responsive to recent changes in the data, such as trend
and seasonality. Which is the most common case of electronic industries due to festival
discounts, seasonal sales and product release cycle.
● Justification for the choice of method: For short term forecasting of the period of 6
months for the given data, The Weighted Moving Average is a suitable choice. As this
method is ideal when recent data is expected to have a more significant impact on future
demand. It is very useful when we want our data to be more responsive to recent
changes such as seasonality or sudden shifts in demand, given that the data of demand
provided is sensitive to changes. In this method, one of the assumptions being made is
that the historical sales data accurately represents the future demand patterns. This
assumes that factors influencing demand such seasonality or market dynamics will
remain constant for the short term period and any other external factor or events that
could affect demand significantly, should be considered separately in forecasting.
Task 2
● The formula used for forecasting demand: ( (Dec Actual)*50%) + ( (Nov Actual)*30%) + (
(Oct Actual)*20%) = January Forecast
● The demand forecast for the next six months is shown in the table below
Month Smartphones (Forecast) Laptops (Forecast) Smartwatches (Forecast)
Jan 2110 1055 359
Feb 2145 894 323
Mar 2166 881 318
Apr 2148 939 334
May 2153 995 347
Jun 2154 1065 363
Task 3
● The formula used for calculating forecast errors: (Actual Demand – Forecasted
Demand)*100/Actual Demand.
● The forecast errors for the six months is shown in the table below
Month Smartphones Laptops Smartwatches
Jan -32 -40 -39
Feb -30 -3 -11
Mar -38 0 -27
Apr -23 -2 -56
May -12 1 -74
Jun -6 3 -85
Task 4
● The formula for calculating MAPE: Ʃ l(Actual-Forecast)l / Ʃ Actual
● MAPE for all product categories for the six months:
Month Smartphones Laptops Smartwatches
Jan 32 40 39
Feb 30 3 11
Mar 38 0 27
Apr 23 2 56
May 12 1 74
Jun 6 3 85
MAPE 23 8 49
● The formula for calculating Bias: Total Variances (Forecast – Actual) / Total Actual.51
● Bias for all product categories for the six months:
Month Smartphones Laptops Smartwatches
Jan 510 300 100
Feb 496 29 31
Mar 597 4 68
Apr 398 16 120
May 233 -5 147
Jun 114 -35 167
Total Variations 2347 309 633
Total Actual 10528 5520 1411
BIAS=(Total 22.3% 5.6% 44.9%
Variation/Total
Actual)*100
Task 5
● Recommendations:
○ Recommendation: (Smart Phones)
Refine Forecasting Methods: For smartphones, continue using weighted
moving average as the forecasts are almost accurate and will help in keeping
sufficient stocks without overstocking. However, for laptops and smartwatches,
consider more advance and accurate forecasting techniques.
○ Recommendation: (Laptops)
Demand Collaborate: Work closely with sales and marketing teams to gather
real-time market analysis. This will help in tuning the forecasts according to
consume changing preferences.
Supplier Agility: Work closely with supplier to reduce lead time, especially for
laptops, where demand fluctuates significantly.
○ Recommendation: (Smartwatches)
Safety Stock Planning: the higher fluctuation rate in smartwatches and higher
MAPE, indicates that safety stock should be there in case of stockouts.
Continuous Improvement: Consistently asses and refine forecasting methods
based on actual performance and feedback. Implement a robust S&OP process
to improve overall supply chain process.
In conclusion, all these suggestions will help in improve the supply chain process
thoroughly in accordance with demand fluctuations, reduce lead time and
improve efficiency, where the forecasts of product are not always accurate.
