Page 75 - eBook_Proceedings of the International Conference on Digital Manufacturing V1

Leveraging Machine Learning to Enhance Supply Chain Agility and Strategic
Operational Excellence

Model 1

The initial phase of the machine learning pipeline involved
importing essential Python libraries to facilitate data
manipulation, numerical computations, model development, and
evaluation as shown in Figure 15. Pandas was utilised for
handling datasets in Data Frame structures, enabling efficient data
loading, cleaning, and preprocessing. NumPy supported
numerical operations and array handling, providing the foundation
for mathematical and statistical computations. The train_test_split
function from sklearn. Model selection was employed to partition
the dataset into training and testing subsets, ensuring the model's
performance could be evaluated on unseen data. For model
development, the XGBRegressor class from the XGBoost library
was used to build and train the machine learning model, while the
plot importance function facilitated visualisation of feature
significance. Model performance was assessed using metrics such
as Mean Absolute Error (MAE), Mean Squared Error (MSE), and
R-squared (R²) from sklearn.metrics. Matplotlib's pyplot module
was employed to create visual representations of data distributions
and model performance. The dataset, assumed to be in CSV
format, was loaded into a Pandas Data Frame using the read_csv()
function, setting the stage for subsequent data preprocessing and
model training. By importing these libraries and loading the
dataset, we set up the foundation for the rest of the machine
learning pipeline, which includes data preprocessing, model
development, and evaluation.

59