PROJECT

SUBMISSION

Task 1
METHOD OF FORECASTING:
I suggest adopting a time series forecasting method, more especially the Exponential
Smoothing method, like the Holt-Winters method, to predict demand for TechHub Electronics for
the upcoming six months (a short-term projection). This approach is useful for making
short-term forecasts and is especially suitable for data that shows a trend and seasonality,
which is frequently the case in the electronics sector due to seasonal sales, holidays, and
product release cycles.

JUSTIFICATION FOR THE CHOICE OF METHOD:

Trend: The data also reveals an overall upward trend in sales for all three product categories
over the previous year. In a market that is rapidly changing, the trend component of the
exponential smoothing method is crucial because it can adjust to shifting sales trends.

Data Sensitivity: Because it gives different weights to more recent and older observations, the
exponential smoothing method is well adapted to capture and react to current changes in
demand patterns. As was previously indicated, the given data is sensitive to recent changes.

Assumption: When applying the Exponential Smoothing approach, it is important to make the
assumption that past trends will hold true in the future. This signifies that we anticipate the
seasonality and pattern seen over the previous 12 months to continue for the ensuing six
months. Although this presumption might not always be true in a highly dynamic market like
electronics, it offers a reasonable place to start when forecasting the near future.

• In actuality, I would advise routinely (e.g., monthly) re-evaluating the

forecast and modifying it when new information becomes available.

Furthermore, it's crucial to take into account outside variables like new

product launches, marketing campaigns, or economic shifts that could

affect demand and modify the forecast appropriately.

Exponential Smoothing approach, we would require historical data for at

least three complete seasons (for example, three years). Even though

there is only one year of data available in this scenario, the prediction

can nevertheless offer useful information for inventory management and

short-term planning.

Task 2
The formula used for forecasting demand: ESF : Alpha*Actual
demand+1-Alpha*forecast demand
The demand forecast for the next six months is shown in the table below

Smartwatches
Month Smartphones Laptops (Forecast) (Forecast)
(Forecast)

Jan 2044.92 1023.10 352.48

Feb 2121.44 1061.17 360.74

Mar 2175.01 1087.82 366.52

Apr 2212.51 1106.47 370.56

May 2238.76 1119.53 373.39

Jun 2257.13 1128.67 375.37

• The formula used for calculating forecast errors: Forecasting error = (Actual
demand - forecast demand) * 100 / actual demand

• The forecast errors for the six months is shown in the table below

Month Smartphones Laptops Smartwatches

Jan 444.92 268.10 93.48

Feb 472.44 196.17 68.74

Mar 606.01 210.82 116.52

Apr 462.51 183.47 156.56

May 318.76 119.53 173.39

Jun 217.13 28.67 179.37

Task 4

• The formula for calculating MAPE: MAPE = Σ(|actual – forecast| / |actual|)

• MAPE for all product categories for the six months:

Month Smartphones Laptops Smartwatches

Feb 28.65 22.68 23.54

Mar 38.62 24.04 46.61

Apr 26.43 19.88 73.16

May 16.60 11.95 86.70

Jun 10.64 2.61 91.52

• The formula for calculating Bias: Total Variances (Forecast- Actual)/Total Actual

• Bias for all product categories for the six months:

Month Smartphones Laptops Smartwatches

Jan 0.042 0.0486 0.0662

Feb 0.045 0.0355 0.0487

Mar 0.058 0.0382 0.0825

Apr 0.044 0.0332 0.1109

May 0.030 0.0217 0.1228

Jun 0.021 0.0052 0.1271

Based on the forecast errors and metrics calculated for each product category
(Smartphones, Laptops, and Smartwatches) and the goal of improving the supply
chain planning process, here are three recommendations, one for each product
category:

1. Smartphones:
Recommendation: Diversify Suppliers and Implement
Risk Mitigation
Rationale: It's critical to diversify smartphone component suppliers and forge solid bonds with a
number of vendors given the high demand volatility and potential supply chain interruptions in
the electronics sector. By employing this technique, the danger of single-source dependence will
be reduced, which can result in production delays and shortages in the event of unforeseen
circumstances like natural disasters or geopolitical upheaval.

Actions to Take:
• Identify the crucial parts used in the manufacture of smartphones and evaluate the
concentration of suppliers for each.
• To lessen the impact of regional disruptions, procure essential components from a
number of suppliers, preferably located in several geographic locations.
• Create backup plans and risk-reduction methods, such as buffer stocks and alternate
sourcing choices, to quickly manage supply chain interruptions.

2. Laptops:
Recommendation: Implement Advanced
Inventory Management Techniques
Rationale: To handle demand fluctuations better and reduce lead times for laptops, it's essential
to implement advanced inventory management techniques. Laptops often have longer lead
times due to their complex manufacturing processes, making it challenging to respond quickly to
sudden demand spikes.
Actions to Take:
• Implement Just-In-Time (JIT) inventory management practices to minimize carrying costs
while ensuring product availability.

• Employ demand sensing and demand shaping tools to better predict and shape
customer demand, allowing for more proactive inventory management.
• Optimize safety stock levels by considering factors like seasonality, supplier reliability,
and transportation lead times.

3. Smartwatches:
Recommendation: Enhance Demand Collaboration with
Retailers
Rationale: Smartwatches exhibit relatively stable demand throughout the year compared to
smartphones and laptops. To enhance supply chain performance for smartwatches, it's crucial
to establish better collaboration and communication channels with retailers to align
production with actual demand.

Actions to Take:
• Implement a Vendor-Managed Inventory (VMI) system or Collaborative Planning,
Forecasting, and Replenishment (CPFR) with key retailers to share demand forecasts and
inventory data.
• Regularly review sales data and collaborate with retailers to adjust production plans
based on real-time demand insights.

• Consider offering incentives to retailers for timely and accurate demand information

sharing to improve forecast accuracy and reduce overstocking or understocking issues.