Data Science for Supply Chain Managers Training Course

Course overview

Supply chain managers are expected to explain service failures, inventory exposure, supplier risk and capacity constraints with evidence rather than intuition. Yet operational data is often split across ERP extracts, warehouse systems, transport records and supplier files. This course equips managers to turn those sources into decision-ready analysis: identify the drivers of late orders, distinguish normal demand variation from genuine disruption, quantify safety-stock trade-offs and communicate recommendations that operations, finance and procurement can act on.

Participants apply data science methods to core supply chain decisions, using SQL to prepare operational data, Python to analyse and model it, and Power BI to present findings. They learn demand segmentation, time-series forecasting, inventory policy analysis, ABC-XYZ classification, supplier performance scoring, lead-time variability analysis, network and capacity scenario modelling, and optimisation with constraints. The course also addresses data quality, feature design, model validation, forecast-error measures and the practical governance required before analytical models influence replenishment or sourcing decisions.

Instruction combines expert-led demonstrations with a connected case based on a multi-site distributor facing stockouts, excess inventory and supplier delays. Each participant works with realistic order, inventory, shipment and purchase-order data in Jupyter Notebook, then develops a Power BI decision dashboard. They leave with a documented supply chain analytics capstone: a data model, analytical notebook, forecast or risk model, dashboard, prioritised recommendations and a 90-day implementation plan for their own function.

The programme is designed for experienced supply chain professionals who already understand planning, sourcing, logistics or warehouse operations and now need stronger analytical capability to lead evidence-based improvement.

Course objectives

By the end of this course, participants will be able to:

• Construct a supply chain data model by joining order, inventory, shipment and purchase-order datasets with SQL
• Profile and remediate missing values, duplicate records, outliers and master-data inconsistencies in operational data
• Segment SKUs using ABC-XYZ analysis and translate segment rules into inventory-control actions
• Build and validate demand forecasts using baseline, moving-average and time-series models
• Calculate safety stock, reorder points, service levels and inventory exposure under variable demand and lead times
• Develop supplier scorecards that measure on-time-in-full performance, lead-time variability and quality risk
• Formulate constrained replenishment or allocation scenarios using Google OR-Tools optimisation models
• Deliver a Power BI supply chain dashboard and 90-day analytics implementation plan for a live business decision

Benefits of attending

Target competencies

Who should attend

• Supply Chain Managers — who need to defend inventory, service and capacity decisions with operational evidence
• Demand Planning Managers — who must improve forecast accuracy and convert forecasts into replenishment actions
• Inventory Control Managers — who balance working capital, safety stock and product availability across SKU portfolios
• Procurement and Sourcing Managers — who need fact-based supplier performance and lead-time risk analysis
• Logistics and Distribution Managers — who investigate delivery failures, transport variability and warehouse throughput
• Supply Chain Transformation Leads — who must define practical analytics use cases and adoption plans across functions

Requirements and prerequisites

Participants should have practical experience in at least one supply chain area, such as demand planning, inventory control, procurement, logistics or S&OP, and be comfortable interpreting KPIs including fill rate, lead time, inventory turns and forecast error. Basic spreadsheet skills and confidence working with tables, filters and formulas are essential. The course assumes introductory familiarity with SQL queries and Python variables, data frames or notebooks; participants should be able to read simple code, though they do not need to be software developers. Prior machine-learning, optimisation or Power BI experience is not required. A laptop capable of running a current web browser and local Python environment is needed.

Training methodology

The programme alternates short instructor-led briefings with guided analysis in SQL, Python and Power BI. Participants work through a continuing distributor case, beginning with messy ERP-style extracts and progressing to forecasting, inventory-policy design, supplier analysis and constrained allocation scenarios. Exercises require participants to inspect assumptions, compare model performance and explain operational consequences rather than merely produce charts. Small-group review sessions simulate an S&OP or supply review, where participants challenge recommendations using service, cost and working-capital evidence. The final day is devoted to refining each participant’s capstone and 90-day application plan.

Course outline

Day 1: Supply Chain Analytics Decision Framework

• Supply chain decision hierarchy from execution to network planning
• KPI tree linking service, cost, cash and risk
• Operational data sources across ERP, WMS, TMS and procurement systems
• Analytical problem framing with decision, owner and action definitions
• Descriptive, diagnostic, predictive and prescriptive analytics distinctions
• Unit-of-analysis selection for SKU-location and order-line analysis
• Python and Jupyter Notebook workflow for supply chain datasets

Workshop: Map a distributor’s stockout problem into a decision canvas that defines required data, KPIs, stakeholders and a measurable intervention.

Day 2: Data Preparation and SQL Modelling

• Relational data structures for orders, inventory, suppliers and shipments
• SQL joins for assembling SKU-location performance records
• Date, calendar and fiscal-period transformations
• Data profiling for nulls, duplicates and referential-integrity failures
• Outlier detection in quantities, prices and lead times
• Master-data standardisation for SKU, supplier and location identifiers
• Reproducible data-preparation pipelines in Python

Workshop: Build a clean analytical dataset from fragmented order, inventory and purchase-order extracts and document the data-quality exceptions found.

Day 3: Descriptive Analytics and Segmentation

• Order-fill rate, OTIF and perfect-order metric construction
• Inventory turns, days of supply and ageing analysis
• Pareto analysis of revenue, volume and inventory value
• ABC classification by annual consumption value
• XYZ classification using demand variability
• SKU-location segmentation matrices and policy implications
• Power BI measures and drill-through analysis

Workshop: Create an ABC-XYZ portfolio view and recommend differentiated review frequencies, service targets and replenishment approaches for each segment.

Day 4: Demand Pattern Analysis and Forecasting

• Demand-history cleansing for returns, promotions and stockout-censored sales
• Trend, seasonality, intermittency and lifecycle pattern identification
• Naive and seasonal-naive forecasting baselines
• Moving-average and exponential-smoothing models
• Time-series train-test splits and rolling validation
• MAPE, WAPE, bias and forecast-value-add metrics
• Forecast exception thresholds and planner review queues

Workshop: Build and compare demand forecasts for a SKU portfolio, then select models using WAPE, bias and operational exception rules.

Day 5: Inventory Policy and Service-Level Analysis

• Demand and lead-time variability in inventory policy design
• Cycle stock, safety stock and pipeline inventory calculations
• Reorder point and order-up-to level methods
• Service level, fill rate and stockout-risk distinctions
• Economic order quantity assumptions and limitations
• Multi-echelon inventory considerations for central and regional stock
• Working-capital impact of policy changes

Workshop: Calculate revised safety stock and reorder points for selected SKU-locations and quantify the service-versus-cash trade-off.

Day 6: Supplier and Lead-Time Risk Analytics

• Purchase-order lifecycle and promised-versus-actual date analysis
• On-time-in-full supplier performance calculations
• Lead-time distribution, variability and percentile analysis
• Supplier quality, quantity and price variance measures
• Supplier scorecard weighting and threshold design
• Risk segmentation by spend, criticality and substitutability
• Corrective-action triggers and sourcing escalation rules

Workshop: Develop a supplier scorecard that identifies high-impact delivery risks and proposes targeted corrective actions for the sourcing team.

Day 7: Logistics and Warehouse Performance Analysis

• Shipment milestone data and transit-time analysis
• Lane-level carrier performance and delay distributions
• Warehouse throughput, dwell time and pick-cycle metrics
• Order backlog ageing and fulfilment bottleneck detection
• Capacity utilisation and queueing indicators
• Root-cause analysis using drill-down and cohort comparisons
• Power BI operational alert and exception design

Workshop: Investigate a late-delivery spike using shipment and warehouse data, then produce a root-cause briefing with operational countermeasures.

Day 8: Optimisation and Scenario Modelling

• Decision variables, objective functions and operational constraints
• Linear programming concepts for supply allocation
• Google OR-Tools model structure and solver workflow
• Constrained allocation under limited supply
• Capacity and minimum-order-quantity constraints
• What-if scenarios for service, cost and inventory trade-offs
• Sensitivity analysis and feasibility interpretation

Workshop: Use Google OR-Tools to allocate constrained supply across customers or locations and compare the service and margin consequences of alternative rules.

Day 9: Decision Dashboards and Model Governance

• Executive dashboard design for supply chain decisions
• Power BI star schemas, measures and refresh logic
• Forecast and inventory exception dashboard layouts
• Model explainability for operational stakeholders
• Data lineage, ownership and access-control requirements
• Model monitoring for drift, bias and performance deterioration
• Human-in-the-loop approval controls for planning decisions

Workshop: Build a Power BI dashboard that combines forecast accuracy, inventory exposure and supplier risk into an actionable weekly review pack.

Day 10: Capstone Recommendation and Implementation Planning

• Capstone analytical narrative from business question to recommendation
• Validation of assumptions and decision thresholds
• Benefit sizing for service, cash and risk outcomes
• Stakeholder mapping for planners, buyers, operations and finance
• Analytics operating model and role definition
• 90-day implementation roadmap and pilot design
• Executive presentation and challenge-session techniques

Workshop: Present a completed supply chain analytics capstone containing the data model, notebook, dashboard, recommendations and 90-day implementation plan.

Tools & standards covered

Python, Jupyter Notebook, Microsoft Power BI, Google OR-Tools

A typical training day

08:30 – 10:30	First session
10:30 – 10:45	Refreshment break
10:45 – 12:30	Second session
12:30 – 13:30	Lunch and networking
13:30 – 15:00	Third session
15:00 – 15:15	Refreshment break
15:15 – 16:30	Workshop and daily review

Live online deliveries follow the same structure in the East Africa Time zone, with shorter screen blocks and longer breaks.

What the fee includes

How you can take this course

Certification

Participants who complete the full five days receive the Skillset Development Certificate of Completion, stating the course title, course code, dates and delivery format — suitable for professional-development records and employer reimbursement.

Frequently asked questions