Description

This curriculum spans the technical and operational rigor of a multi-workshop inventory optimisation initiative, matching the depth required for configuring enterprise forecasting systems and supporting ongoing service parts network reviews.

Module 1: Foundations of Service Parts Inventory Behavior

Selecting between intermittent, lumpy, and regular demand classification models based on historical transaction frequency and variance.
Determining appropriate historical lookback periods for demand analysis given product lifecycle stage and obsolescence risk.
Deciding whether to include or exclude outlier demand events (e.g., one-time field retrofits) in baseline statistical models.
Mapping service parts to equipment serial numbers and installed base data to validate demand causality.
Establishing rules for handling zero-demand periods in forecasting when parts have long dormancy but sudden failure spikes.
Integrating engineering change notifications into demand pattern analysis to preempt obsolete stock accumulation.

Module 2: Demand Forecasting for Low-Volume Parts

Choosing between Croston’s method, SBA, and TSB for intermittent demand based on bias and service level performance in backtesting.
Adjusting forecast outputs manually when known future fleet retirements or regulatory changes invalidate statistical trends.
Implementing hierarchical forecasting reconciliation when parts serve multiple equipment platforms with shared components.
Setting thresholds for when to switch from statistical forecasting to expert judgment due to data sparsity.
Validating forecast accuracy using period-error metrics that reflect business impact, such as days of inventory deviation.
Managing forecast inputs during parts consolidation or substitution programs that alter historical demand streams.

Module 3: Lead Time Variability and Supply Constraints

Calculating effective lead time by incorporating supplier reliability data, customs delays, and internal receiving bottlenecks.
Adjusting safety stock targets when dual-sourcing transitions to single-source due to supplier rationalization.
Accounting for internal transfer lead times between regional distribution centers in multi-echelon networks.
Updating lead time distributions when suppliers move production offshore, increasing both duration and variance.
Handling planned supply interruptions (e.g., factory shutdowns) by pre-positioning stock without triggering excess alerts.
Integrating supplier quality failure rates into lead time risk when rework or rejection cycles extend replenishment duration.

Module 4: Service Level Definition and Segmentation

Assigning differential service levels to parts based on equipment criticality, downtime cost, and contractual SLAs.
Resolving conflicts between field service KPIs (e.g., mean time to repair) and inventory turnover targets.
Defining service level as line-item fill rate versus order fill rate, and managing the operational implications.
Segmenting inventory using ABC-XYZ analysis while adjusting for low-volume high-criticality exceptions.
Revising service level targets quarterly based on actual equipment uptime data and customer penalty clauses.
Managing stakeholder expectations when high-cost, low-demand parts cannot meet aggressive service level goals.

Module 5: Safety Stock Calculation and Model Selection

Selecting between normal, gamma, and negative binomial distributions for demand variability based on empirical fit.
Implementing simulation-based safety stock models when analytical methods fail due to non-stationary demand.
Adjusting safety factor multipliers when demand and lead time distributions are correlated, not independent.
Validating model outputs against historical stockout events to calibrate confidence intervals.
Applying bootstrapping techniques to estimate safety stock when historical data spans fewer than 24 months.
Documenting model assumptions and limitations for audit and handover to supply chain analysts.

Module 6: Multi-Echelon Inventory Optimization

Determining optimal stocking locations for common parts across central warehouses, regional hubs, and forward depots.
Setting push-pull boundaries in the network based on replenishment lead time and demand volatility.
Allocating safety stock across echelons using METRIC or approximate optimization when exact methods are computationally infeasible.
Managing lateral transshipments between depots under formal borrowing policies with replenishment recovery rules.
Updating echelon-level targets when opening or closing service centers in response to market shifts.
Integrating repair cycle times and return forecasts into safety stock calculations for reusable service parts.

Module 7: Governance, Review Cycles, and Exception Management

Establishing review frequency for safety stock parameters based on part criticality and volatility bands.
Designing exception reports that flag parts with sustained overstock or chronic stockouts for root cause analysis.
Defining approval workflows for manual overrides to algorithmically generated safety stock values.
Integrating new product introduction (NPI) forecasts into safety stock planning with conservative initial settings.
Conducting quarterly inventory health assessments that include obsolescence risk and carrying cost per service level tier.
Aligning safety stock policies with financial close processes to ensure accurate balance sheet provisioning.

Module 8: System Integration and Tool Configuration

Configuring ERP safety stock parameters to reflect statistical outputs while preserving audit trails for manual adjustments.
Mapping master data fields between CMMS, ERP, and forecasting tools to ensure consistent part and location identifiers.
Setting data refresh schedules that align demand updates with procurement cycle cutoffs.
Validating system-generated reorder points against manual calculations during parallel run periods.
Designing user roles and access controls for safety stock parameter maintenance to prevent unauthorized changes.
Automating alerts for parameter drift when actual performance deviates beyond tolerance from model assumptions.