Manage carrier relationships, demand forecasting, inventory replenishment, freight exceptions, production scheduling, and returns across the full supply chain lifecycle.
End-to-end supply chain operations covering carrier management, inventory planning, exception handling, production scheduling, and reverse logistics.
Source, vet, and negotiate with freight carriers across truckload (TL), LTL, intermodal, and brokerage.
Every freight rate has components to negotiate independently — bundling obscures overpayment:
Carrier scorecard — track 5 metrics, not 20:
| Metric | Target | Red Flag |
|---|---|---|
| On-time delivery (OTD) |
| ≥95% |
| <90% |
| Tender acceptance rate | ≥90% (primary) | <80% |
| Claims ratio (% of spend) | <0.5% | >1.0% |
| Invoice accuracy | ≥97% | <93% |
| Tender-to-pickup time | Within 2 hrs (FTL) | Consistent late pickup |
FMCSA compliance: Verify active MC authority, insurance minimums ($1M — above FMCSA's $750K minimum), safety rating (never Unsatisfactory), and broker bond ($75K) before first load and quarterly thereafter.
Portfolio strategy: 60–70% asset carriers, 20–30% brokers, 5–15% specialty. Build 3-deep routing guides for lanes with >2 loads/week. No single carrier exceeds 40% of any critical lane.
RFP process (8–12 weeks): Analyze 12 months of lane data → lane-level bid packages (not portfolio bids) → evaluate on 40–50% cost / 25–30% service history / 15–20% capacity commitment → award in waves → 30-day parallel period.
Carrier exit triggers (after documented corrective action): OTD <85% for 60 days, tender acceptance <70% for 30 days, claims ratio >2% for 90 days, FMCSA lapse, double-brokering confirmed.
Translate commercial intent into executable purchase orders while minimizing stockouts and excess inventory.
Forecasting method selection:
| Demand Pattern | Primary Method | Review Trigger |
|---|---|---|
| Stable, high-volume | Weighted moving average (4–8 weeks) | WMAPE >25% for 4 weeks |
| Trending | Holt's double exponential smoothing | Tracking signal exceeds ±4 |
| Seasonal | Holt-Winters (multiplicative) | Season correlation <0.7 |
| Intermittent (>30% zero periods) | Croston's method | Mean inter-demand interval shifts >30% |
| Promotion-driven | Causal regression (baseline + lift layer) | Post-promo actuals deviate >40% |
Safety stock: SS = Z × σ_d × √(LT + RP). For lead time variability: SS = Z × √(LT_avg × σ_d² + d_avg² × σ_LT²). Service level targets: A-items 95–97.5%, B-items 95%, C-items 90–92%, CZ 85%.
ABC/XYZ classification: A = top 20% SKUs driving 80% revenue. X = CV <0.5 (predictable). AX: automated replenishment with tight SS. AZ: human review every cycle. CZ: candidate for discontinuation.
Promotional lift: Strip promo volume from baseline before fitting models. Typical lifts: 15–40% for price reduction only, 80–200% for TPR + display + circular, 300–500%+ for doorbusters. Always model post-promo dip (default: 40% of incremental lift, concentrated in week 1).
Forecast accuracy metrics: WMAPE <25% target, >35% red flag. Bias ±5% healthy, >±10% for 4 weeks signals structural model problem. Tracking signal exceeds ±4 → re-parameterize or switch methods.
Escalation triggers: Projected stockout on A-item within 7 days → alert within 4 hours. Vendor lead time increase >25% → notify within 1 business day. Promo forecast miss >40% → debrief within 1 week.
Resolve freight exceptions — delays, damages, losses, and carrier disputes — quickly while protecting financial interests.
Exception taxonomy: Delay (transit) ~40% of all exceptions. Damage (visible, concealed, temperature). Shortage / overage. Refused delivery. Lost (full or partial shipment). Contaminated.
Claims process:
Eat-the-cost vs. fight-the-claim thresholds:
$10,000 → VP-level awareness, reject <90%, legal review if denied
Priority sequencing (multiple active exceptions): (1) Safety/regulatory, (2) Production shutdown risk, (3) Perishable <48 hours shelf life, (4) Highest financial impact by customer tier, (5) Oldest unresolved exception.
Exception rate targets: <25 per 1,000 shipments (red flag: >40). Mean resolution time <72 hours (red flag: >120). Financial recovery rate >75% (red flag: <50%).
Sequence jobs, balance lines, optimize changeovers, and resolve disruptions to maximize constraint throughput.
Theory of Constraints (DBR): Identify the drum (constraint — work centre with highest load/capacity ratio, typically >85%). Buffer time = 50% of production lead time for the constraint operation. The rope limits new work release to the constraint's rate. A minute lost at the constraint is a minute lost for the entire plant.
Buffer zones: Green (<33% consumed) = well-protected. Yellow (33–67%) = expedite upstream. Red (>67%) = immediate management action, potential overtime upstream.
Job priority sequencing:
Changeover optimization: Build setup matrix (changeover time and cost for each product pair). Apply nearest-neighbour heuristic then 2-opt swaps. Validate against due dates — due date compliance trumps changeover optimization.
SMED methodology: Classify setup elements as internal (machine stopped) or external (machine running). Convert internal to external where possible. Typical result: 40–60% setup time reduction from reclassification alone.
Disruption response (within 30 minutes): Assess impact window → freeze jobs in-process or within 2 hours of start → re-sequence unlocked jobs using priority framework → communicate revised schedule → lock for minimum 4 hours.
OEE: Availability × Performance × Quality. World-class: 85%+. Typical discrete manufacturing: 55–65%. A 2% yield improvement at the constraint equals a 2% capacity expansion.
Key performance targets:
| Metric | Target | Red Flag |
|---|---|---|
| Schedule adherence (±1 hr) | >90% | <80% |
| On-time delivery | >95% | <90% |
| OEE at constraint | >75% | <65% |
| Constraint utilization | >85% | <75% |
| Unplanned downtime | <5% | >10% |
Authorize, inspect, disposition, and recover value from returned goods while detecting fraud and managing vendor recovery.
Condition grading:
Disposition routing by category:
| Category | Grade A | Grade B | Grade C | Grade D |
|---|---|---|---|---|
| Consumer Electronics | Restock (test first) | Open box / Renewed | Refurb if ROI >40%, else liquidate | Parts or e-waste |
| Apparel | Restock if tags on | Repackage / outlet | Liquidate by weight | Textile recycling |
| Health & Beauty | Restock if sealed | Destroy | Destroy | Destroy |
Fraud scoring — flag for review at 65+, hold refund at 80+:
Vendor recovery: Pursue claims >$500. Batch claims $200–$500. Offset claims <$200 against next PO. Raise threshold to $1,000 for overseas vendors. RTV claim window is typically 90 days from receipt — do not let eligible product sit past this window.
Key metrics:
| Metric | Target | Red Flag |
|---|---|---|
| Processing time (receipt to refund) | <48 hours | >96 hours |
| Restock rate | >45% | <30% |
| Fraud detection rate | >80% | <60% |
| Vendor recovery rate | >70% | <45% |
| Cost per return processed | <$8.00 | >$15.00 |
Automatic escalation: Return value >$5,000 → supervisor approval before refund. Fraud score ≥80 → hold and route to fraud review immediately. Recalled product identified → do not process as standard return.