Business Process Modelling
Autor: betaj badshah • December 4, 2016 • Case Study • 477 Words (2 Pages) • 915 Views
Business Process Modelling
Assignment#1
Muhammad Waqar Majeed
19110342
(a)
Problem
Yankee fork and hoe company is a leading producer of garden tool. The situation is a such that they face fierce competition in garden tools market. They are facing some problems recently with regards to their long-term customer. They are unable to ship certain goods on time.
Forecasting method
Forecasting of bow rake is “informal” according to Phil Stanton. The marketing manager, Ron Adams, meet with the regional managers and they go over shipping data of past year and take into account variable factors that can affect forecast figure such as; changes in economy, last year shortages and anticipated promotions. They do couple of meetings before arriving at monthly forecast figures for the next year. Ron admits that despite considerable time and effort, they still face customer problems.
This forecast is then forwarded to production head, Phil Stanton. He feels that forecasting figures prepared by marketing departments are “inflated”. They are in a long-term purchase relationship for steel and its storage is also expensive. For all these reason, Phil decides to reduce the figure by 10%. These modified figures are then used to generate monthly final assembly schedule for next year.
Recommendations
- More frequent meetings between regional managers as well as different departments
- Changes in the final forecast should not be made by single department but rather should involve all the concerned departments
- The primary data they use is past shipping data. A more appropriate data to be used in calculating the forecast may be the actual historical demand.
- They can use weighted moving average to calculate the forecast. It is not only convenient but also smooths out the random spikes in the data.
- They do not take into account any mathematical models despite having extensive historical data
- Reliance on judgements and opinions (naïve approach) which is not a very good idea not
(b)
Since there are random spikes in demand month to month, we should use forecasting method that not only smooths out these spikes but also make use of the historical trend month wise. Therefore, I would use Weighted Moving Average to calculate the year 5 forecasted demand.
We could also have also used exponential smoothing technique here but the reason I am not using it here is because:
- We have extensive past data available easily for each month
- The market conditions for these tools are stagnant in the sense that there is room for product improvisation. As same tools are being used for the past 30 years and so past data would be appropriate representative of the future forecast
- Actual demand is needed to calculate forecasts according to this method which we don’t have.
- No predictive variables are mentioned and so using associative models are of no use
Month | Year1 | year 2 | Year 3 | year4 | year 5( weighted moving average) |
1 | 55,220 | 39,875 | 32,180 | 62,377 | 48101.8 |
2 | 57,350 | 64,128 | 38,600 | 66,501 | 56741 |
3 | 15,445 | 47,653 | 25,020 | 31,404 | 31142.7 |
4 | 27,776 | 43,050 | 51,300 | 36,504 | 41379.2 |
5 | 21,408 | 39,359 | 31,790 | 16,888 | 26304.8 |
6 | 17,118 | 10,317 | 32,100 | 18,909 | 20968.8 |
7 | 18,028 | 45,194 | 59,832 | 35,500 | 42991.2 |
8 | 19,883 | 46,530 | 30,740 | 51,250 | 41016.3 |
9 | 15,796 | 22,105 | 47,800 | 34,443 | 34117.8 |
10 | 53,665 | 41,350 | 73,890 | 68,088 | 63038.7 |
11 | 83,269 | 46,024 | 60,202 | 68,175 | 62862.3 |
12 | 72,991 | 41,856 | 55,200 | 61,100 | 56670.3 |
Year1 | year 2 | Year 3 | year4 | ||
weights assigned according to time | 0.1 | 0.2 | 0.3 | 0.4 | (weights obtained from the book for ease of use) |
Formula used= | SUMPRODUCT(B2:E2,$B$16:$E$16) |
...