Page 79 - eBook_Proceedings of the International Conference on Digital Manufacturing V1
P. 79
Leveraging Machine Learning to Enhance Supply Chain Agility and Strategic
Operational Excellence
Figure 19: Line Plot between Actual vs. Predicted Demand
From the above findings, from the R2 value and the graphs
between “Actual vs Predicted Products Sold”, the value of R2
have increased from −0.428 to 0.044 when shifted the data set
from XGBoost to Random Forest Modelling Technique and the
values are somewhat close to the actual line and sorted, as shown
in Figure 20. It is evident that this model is slightly better than the
previous one. . However, further iterations are needed to be done
either in further refining the data or moving towards another
machine learning technique to have better and more accurate
results.
Figure 20: Scatter Plot between Actual vs Predicted & Number
Products Sold
63
