Advanced Supply Chain Management – What Error Analysis Will You Do So As To Determine The Accuracy Of The Forecast You Are Making
What error analysis will you do so as to determine the accuracy of the forecast you are making.
What error analysis will you do so as to determine the accuracy of the forecast you are making.
Advanced Supply Chain Management
1. As a Supply Chain department head, you are stocking large inventory to improve customer service level. However, there is a high pressure from the management to reduce inventory cost. Explain with examples how you will plan for inventory optimization without affecting customer service.
2. As a production planning head you are required to decide on a suitable aggregate plan for an export order. The demand forecasted are shown below, for next 6 periods. You have with you at present 1050 workmen and each can produce 1 unit per month (all months have same working days). The beginning inventory is zero. The production cost is $100 per unit and includes material and labor cost. If you want to increase the workforce hiring cost would be $30 per workman and to reduce the workforce the layoff cost would be $70 per workman. Inventory holding cost is $20/unit per month and backorder will incur $50/unit/month. Make Chase and Level strategy and explain with reason which strategy you will recommend.
| Month | 1 | 2 | 3 | 4 | 5 | 6 |
| Demand | 600 | 900 | 1200 | 2000 | 1400 | 800 |
3. You are doing production planning and the performance of your organization depends on the accuracy of the demand forecast you make for it.
a. Explain the six-step approach you may take to ensure that your will conduct an effective forecast.
b. What error analysis will you do so as to determine the accuracy of the forecast you are making.
Answer:
To determine the accuracy of the forecast in production planning, it is essential to conduct a comprehensive forecast error analysis. This process involves comparing forecasted demand against actual demand and measuring the deviation between the two. Forecast accuracy directly impacts inventory management, customer satisfaction, and overall operational efficiency. Several statistical tools and metrics are widely used for this purpose, which help identify not only the magnitude but also the direction of errors.
1. Mean Absolute Deviation (MAD)
The Mean Absolute Deviation measures the average absolute difference between the forecasted and actual values. It provides a straightforward understanding of forecast accuracy in absolute terms without considering the direction of the error.
A lower MAD indicates higher forecasting accuracy. It is particularly useful when the magnitude of errors needs to be communicated in easily understandable terms (Makridakis, Spiliotis, & Assimakopoulos, 2018).
2. Mean Squared Error (MSE) and Root Mean Squared Error (RMSE)
The Mean Squared Error (MSE) takes the square of each forecast error before averaging, giving more weight to larger errors. The Root Mean Squared Error (RMSE) is the square root of MSE and is in the same unit as the original data.
These metrics are effective in penalizing large deviations more severely, making them suitable for identifying volatility in forecasts (Hyndman & Athanasopoulos, 2018).
3. Mean Absolute Percentage Error (MAPE)
The MAPE expresses forecast error as a percentage, making it easier to interpret across different scales of data.
This is particularly useful in a business setting for communicating accuracy to stakeholders, but it can be problematic when actual values are close to zero (Hyndman & Koehler, 2006).
4. Tracking Signal
A Tracking Signal helps monitor the forecast bias by assessing whether errors are consistently positive or negative. It is calculated by dividing the cumulative forecast error by the MAD:
A tracking signal outside a pre-set control limit (usually between -4 and +4) indicates a biased forecast that needs adjustment (Chase, 2016).
5. Bias Analysis
Forecast bias occurs when forecasts are systematically overestimated or underestimated. It is measured using the mean forecast error (MFE):
A positive MFE indicates over-forecasting, while a negative MFE shows under-forecasting. Identifying bias helps improve the forecast model over time.
Conclusion
To ensure high-performance production planning, applying a combination of these error analysis techniques is recommended. MAD and MAPE provide interpretability, RMSE highlights large errors, and tracking signal and bias analysis help adjust forecast strategies. Regular error analysis leads to continuous improvement in forecasting, which enhances operational efficiency and reduces costs.
References
Chase, C. W. (2016). Demand-driven forecasting: A structured approach to forecasting (2nd ed.). Wiley.
Hyndman, R. J., & Athanasopoulos, G. (2018). Forecasting: Principles and practice (2nd ed.). OTexts.
Hyndman, R. J., & Koehler, A. B. (2006). Another look at measures of forecast accuracy. International Journal of Forecasting, 22(4), 679-688.
Makridakis, S., Spiliotis, E., & Assimakopoulos, V. (2018). Statistical and Machine Learning forecasting methods: Concerns and ways forward. PLOS ONE, 13(3), e0194889. https://doi.org/10.1371/journal.pone.0194889
NMIMS SOLVED ASSIGNMENT JUNE 2020,
NMIMS SOLVED ASSIGNMENT SOLUTION,
NMIMS SOLVED ANSWERSHEET JUNE 2020
MBA Solved Assignment Solutions
Project Report & Thesis
Contact us: – PRAKASH
Mobile: – +91- 9741410271 / 08722788493
Email: – smu.assignment@gmail.com