Artificial Intelligence for Supply Chains and Logistics Course

Lesson 1 • AI Maturity in Logistics Organisations

Presents a maturity model from manual operations to autonomous supply chains. Students assess their organisation's current AI readiness level.

Lesson 2 • Supply Chain Architecture Overview

Covers end-to-end supply chain flows: procurement, production, distribution, and returns. Establishes the operational context AI tools must address.

Lesson 3 • Core AI Concepts for Practitioners

Introduces machine learning, deep learning, and rule-based AI without heavy maths. Connects each paradigm to practical supply chain use cases.

Lesson 1 • Data Governance and Privacy

Establishes policies for data ownership, access control, and compliance with privacy regulations. Governance prevents legal and operational risk.

Lesson 2 • Supply Chain Data Sources and Types

Maps transactional, sensor, and market data to supply chain decisions. Establishes the data inventory needed before any AI project begins.

Lesson 3 • Building a Logistics Data Pipeline

Integrates ingestion, transformation, and storage into a repeatable pipeline. Students design a pipeline architecture for a chosen logistics scenario.

Lesson 4 • Feature Engineering for Logistics Models

Transforms raw supply chain variables into predictive features. Strong features reduce model complexity and improve interpretability.

Lesson 1 • Probabilistic and Hierarchical Forecasting

Generates prediction intervals and reconciles forecasts across product hierarchies. Probabilistic outputs enable better safety stock decisions.

Lesson 2 • Time-Series Deep Learning Models

Introduces LSTM and Transformer-based models for sequential demand data. Deep learning captures long-range dependencies that tree models miss.

Lesson 3 • Forecasting Fundamentals and Baselines

Reviews classical forecasting methods as performance baselines. Understanding baselines is essential for measuring ML model improvement.

Lesson 4 • Forecast Model Deployment and Monitoring

Covers model packaging, API integration, and drift detection in production. Deployed forecasts must be monitored to maintain accuracy over time.

Lesson 1 • Multi-Echelon Inventory Management

Extends single-location models to networked warehouse systems. AI coordinates stock across echelons to minimise total holding and shortage costs.

Lesson 2 • Optimisation Algorithms for Inventory

Applies linear programming, genetic algorithms, and simulated annealing to stock decisions. Each method suits different problem scales and constraints.

Lesson 3 • Reinforcement Learning for Replenishment

Models inventory replenishment as a Markov decision process solved by RL agents. RL adapts policies dynamically as demand patterns shift.

Lesson 4 • Inventory Theory and AI Opportunity

Reviews EOQ, safety stock, and reorder point models as the baseline. Identifies where AI adds value beyond classical inventory formulas.

Lesson 1 • Last-Mile Delivery Optimisation

Addresses the high-cost, high-complexity final delivery segment with AI tools. Covers crowd-sourcing, locker networks, and drone routing models.

Lesson 2 • ML-Enhanced Vehicle Routing

Integrates ML predictions of travel time and demand into routing decisions. Predictive inputs reduce route plan deviations and fuel costs.

Lesson 3 • Classical and Heuristic Routing Algorithms

Covers Dijkstra, Clarke-Wright savings, and nearest-neighbour heuristics for routing. Heuristics provide fast, near-optimal solutions for large fleets.

Lesson 4 • Real-Time Fleet Management with AI

Uses live telematics and AI to reroute vehicles and manage exceptions dynamically. Real-time adaptation reduces delays and improves customer satisfaction.

Lesson 1 • Robotics and Autonomous Mobile Robots

Covers AMR navigation, task assignment, and human-robot collaboration in warehouses. AI orchestrates robot fleets to maximise throughput and safety.

Lesson 2 • AI-Driven Slotting and Layout Optimisation

Uses clustering and optimisation to assign SKUs to storage locations. Optimal slotting reduces travel distance and picker fatigue significantly.

Lesson 3 • Computer Vision in Warehouse Operations

Applies object detection and OCR to automate receiving, quality checks, and inventory counts. Vision systems reduce manual errors and processing time.

Lesson 4 • Warehouse Process Mapping for AI

Identifies receiving, putaway, picking, packing, and shipping as AI intervention points. Process mapping reveals where automation delivers the highest ROI.

Lesson 1 • Continuous Risk Monitoring Systems

Builds automated dashboards that ingest live signals and update risk scores in real time. Continuous monitoring replaces periodic reviews with always-on intelligence.

Lesson 2 • AI-Based Disruption Detection

Uses anomaly detection and NLP on news and social data to flag emerging risks. Early detection shortens response time and reduces disruption severity.

Lesson 3 • Resilience Strategy Design with AI

Evaluates dual sourcing, nearshoring, and inventory buffers using AI optimisation. AI identifies the lowest-cost resilience configuration for each risk profile.

Lesson 4 • Scenario Simulation and Stress Testing

Runs Monte Carlo and agent-based simulations to quantify disruption impacts. Simulation outputs guide contingency planning and buffer stock decisions.

Lesson 1 • Change Management for AI Adoption

Addresses workforce resistance, reskilling needs, and cultural shifts during AI rollout. Effective change management is the top predictor of AI project success.

Lesson 2 • Building the AI Business Case

Quantifies cost savings, revenue uplift, and risk reduction from AI investments. A rigorous business case secures executive sponsorship and budget approval.

Lesson 3 • Measuring and Scaling AI Value

Tracks AI ROI through defined KPIs and scales successful pilots across the enterprise. Scaling requires standardised platforms, governance, and centre-of-excellence models.

Lesson 4 • AI Roadmap and Prioritisation

Sequences AI initiatives by value, feasibility, and strategic fit using a scoring matrix. A phased roadmap prevents resource overload and builds momentum.