Page 70 - eBook_Proceedings of the International Conference on Digital Manufacturing V1

Proceedings of the International Conference on Digital Manufacturing –
Volume 1

(d) Inventory Management
ML can optimise inventory levels, reduce stockouts, and
minimise holding costs by predicting demand patterns
and supply chain disruptions (Sucky, 2009). Inventory
management techniques include Economic Order
Quantity (EOQ), Just-in-Time (JIT), and safety stock
management. According to the provided document, EOQ
analysis allows organisations to calculate the ideal
ordering quantity to minimise total inventory expenses,
while JIT inventory management targets inventory levels
by only ordering materials and products when they are
required for production.
(e) Risk Management
ML can identify and assess supply chain risks, such as
supplier failures and disruptions, enabling proactive risk
mitigation strategies (Faisal et al., 2007). Risk
management strategies include supplier diversification,
safety stock competence, and long-term relationship
investments.

Supply Chain Resilience and Visibility

The effectiveness of ML-based supply chain solutions hinges on
rigorous data preprocessing, including cleaning, transformation,
and integration of data, as well as careful feature engineering to
extract relevant insights (Domingos, 2012). Model validation
techniques, like cross-validation and holdout testing, are crucial to
ensure model generalisability. The integration of ML into SCM
enables transformative improvements in demand forecasting, cost
reduction, and operational agility, fostering resilient and adaptive
supply chain infrastructures. These strategies involve optimising
transportation routes, inventory levels, and overall network design
through advanced analytical models. Machine learning effectively
mitigates the bullwhip effect by enhancing demand forecasting
accuracy and improving collaboration and information sharing
across departments and supply chain partners.

54