Coursera Supply Chain Planning Week 4 Assignment Answers

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 6

Supply Chain Planning

Coursera Supply Chain Planning Week 4 Assignment Answers

Learning Updates 5/20/21 Free File


Coursera Supply Chain Planning
Week 4 Peer-Graded Assignment Answers
100% Correct Answers
By Learning Updates

Project Title * Supply Chain Planning

PROMPT

For each one of the four products, what patterns do you see and
what does it mean for the forecast?

Ans.

For analyzing the data, first we plotted the demand against the time
scale to have a better understanding. Once the plots are made, it can
be seen from the plots for Product A. the demand peaks every 3rd
day of the 2nd month. So, for example there is a peak on 3rd January
followed by a peak on 3rd March and then again on 3rd of May and
so on. We can conclude that demand here can be considered as
seasonal. In other words, forecast can be considered into a monthly
bucket. The formula for this case can be Fs = D_(t-2) since the peaks
are based on every alternate 2 months. For Product B, there is no
definite pattern in the chart, except that the demand peaks on the
3rd of march 2011 and 3rd of January 2013. Although there are
occasional high demands, the pattern does not give us a complete
picture of what can be anticipated in the coming day/month. As a
result, it will be best to apply Moving Average Forecast system in this
case. Sub setting the data and averaging it with shorter time periods
1|Page
Supply Chain Planning Assignment Answers
will make the forecast system more reactive. For Product C, the
graph shows the demand increases with time almost at a constant
race (say T). The formula for forecast that can be used here is F_t = T
where T signifies the growth in Demand. For Product D, the graph
shows a complex demand growth over time. A versatile method will
be required to generate the forecast model. Hence. the Exponential
Smoothing method would be the best fit for this demand. The model
will give us the flexibility to make it a cumulative or naive model
based on demand over time.

PROMPT

Which forecasting methods did you try for each product?

Why did you choose those forecasting methods?

Ans.

As per above discussion, since product A and Cs demand displays a


monthly bucket chart, we can utilize the Naive forecast system or
Cumulative Mean forecast system. For Product C we just have to
keep in mind the increasing rate of demand and take into the
consideration of variable T.

For Product B, the Moving Average Forecast system has been used.
Choosing our N to be small, we can make our forecast model more
sensitive and reactive as it can correctly predict the future demand
by averaging the subsets of data in a smaller time frame. The perfect
value of N can be determined by continuously observing the Forecast
Model and overseeing the Mean Squared Error (MSE) value. Lower
the value of the MSE, higher the accuracy of this model. For Product

2|Page
Supply Chain Planning Assignment Answers
D, the Exponential Smoothing Forecast model has been used. A value
of alpha has been determined to make sure the MSE is the lowest
hence making the model most accurate in correctly predicting the
forecast for the next day. If required, the value of alpha can be made
large to give more weightage to demand when demand is low, so
that the model acts like a Naïve model and when demand is high, we
can give more weightage to forecast to make it more cumulative.

PROMPT

What is your conclusion of the results? Do the forecasting methods


work well?

Do you have any thoughts on how could you possibly improve them?

Ans.

After generating the models, the forecast was compared to the last
recorded demand and the forecast pattern was compared to the
demand pattern over time. The models worked very well and for
most cases showed the correct forecast for demand provided that
values of N and alpha were correct. To keep improving on the
models, more data will be required to be recorded and continue to
be monitored. Seeing the behavior values of N and Alpha can be
adjusted to get the lowest MSE and more accurate forecast model.

It can also be interesting to apply different type of models on each


product and see the forecast generated from each model for each
product. This will give us a better representation of which model is
most accurate.

3|Page
Supply Chain Planning Assignment Answers
PROMPT

Now that you have forecasts for these products, what next?

Please outline the next steps in how you would use the forecast to
create a Supply Chain Plan.

Ans.

Now that forecasts have been generated, it will be important to link


it with our operations and transportations of these products. We can
determine the inventory we can keep and how much of product we
need to generate. Following that we can determine our safety stock,
EBQ and EOQ. Transportation of these products from manufacturing
site to warehouse can also be arranged on seeing the forecast and
demand charts. This will provide the warehouse with right stock at
the right time period. If not, it can at least provide the warehouse
with some extra products in case demand warehouse with some
extra products in case demand increases soon.

4|Page
Supply Chain Planning Assignment Answers
If You Liked the Effort then Hit ‘Subscribe’ for
more Interesting Videos!

YouTube Channel Link: https://www.youtube.com/channel/UCKXtnMcSrAid34MSjxW17gQ/

Telegram Channel Link: https://t.me/learning_updates

Thank You for Trusting Us!


@LearningUpdates

5|Page
Supply Chain Planning Assignment Answers

You might also like