Showcase: From Shipment to Savings

A complete walkthrough of every jinflow instrument — using real data from Express Europe, a European freight forwarding operation. Every finding, thesis, and verdict shown here was computed from actual shipment data.

The operation

Express Europe is a fast-delivery freight forwarder operating across European routes. Their data comes from CargoWise — shipments, checkpoints, customs declarations, carriers, warehouses, incidents, and inventory snapshots. InterLogic ingests it all through the medallion pipeline and builds a knowledge store.

The numbers: 194,973 shipments. 1,667,442 checkpoints. 194,973 customs declarations. 584,241 shipment items. 7,933 incidents. 25 routes across 5 carriers and 5 warehouses.

Step 1: Entities — the Gold contract

The P-world (pipeline) delivers 12 clean entities:

Entity	Rows	What it represents
Shipments	194,973	Consignments with origin, destination, weight, value
Checkpoints	1,667,442	Location + timestamp events per shipment (picked up, in transit, delivered)
Customs Declarations	194,973	HS codes, declared values, duties per shipment
Shipment Items	584,241	Individual products within each shipment
Carriers	5	Transport companies with SLA agreements
Routes	25	Origin-destination corridors with expected transit times
Warehouses	5	Distribution hubs with capacity and inventory
Customers	500	Shippers and consignees
Incidents	7,933	Delays, damages, losses, customs holds
Inventory Snapshots	180	Periodic warehouse stock positions
Containers	0	Not used — Express Europe operates parcels and pallets only
Route Legs	0	Not used — direct routing model

Step 2: Signals — detect anomalies

5 active signals examine the entities and produce 166,908 findings:

Signal	Findings	Worst	What it detects
probe_delivery_sla_breach	151,157	low	Shipments delivered beyond the carrier’s SLA window
probe_customs_vs_shipment	7,763	medium	Declared customs value diverges from shipment value
probe_hs_classification_mismatch	100	high	Same product declared under different HS tariff codes
probe_carrier_performance	25	low	Carriers with above-threshold incident rates

One perspective aggregates customs-related findings:

Perspective	Entities scored	What it measures
assessment_customs_health	7,863 declarations	Customs compliance combining value gaps and HS mismatches

The worst finding: “Cooking / heating equipment” — declared under 10 different HS codes across shipments. CHF 11.2B in cumulative duty exposure.

Step 3: Theses — is this a pattern?

Four theses evaluate whether the signal findings represent systematic business concerns:

thesis_customs_overpayment — CONFIRMED

“The organization shows systematic customs duty overpayment driven by inconsistent HS classification. The same products are being declared under different tariff codes by different staff, leading to overpaid duties.”

• Status: confirmed (evidence score 0.67)
• Findings: 7,863 across 2 signals
• Money at risk: CHF 258.7B (cumulative duty exposure)
• Evidence: probe_hs_classification_mismatch (primary) + probe_customs_duty_reconciliation (supporting) + probe_customs_vs_shipment (context)

thesis_delivery_reliability — CONFIRMED

“Delivery delays are systemic. A significant share of shipments exceed their SLA window, and carrier performance metrics confirm that the problem is structural — not caused by isolated incidents.”

• Status: confirmed (evidence score 1.0)
• Findings: 151,182 across 2 signals
• Money at risk: CHF 666K
• Evidence: probe_delivery_sla_breach (primary) + probe_carrier_performance (supporting)

thesis_carrier_accountability — CONFIRMED

“Certain carriers consistently fail to meet their SLA commitments. Performance gaps are not route-specific but carrier-specific, indicating a procurement or contract enforcement issue.”

• Status: confirmed (evidence score 1.0)
• Findings: 151,182 across 2 signals
• Money at risk: CHF 666K
• Evidence: probe_carrier_performance (primary) + probe_delivery_sla_breach (supporting)

thesis_customs_value_integrity — CONFIRMED

“Declared customs values systematically diverge from shipment values. The gap is not random — it points to a structural process issue in how commercial invoices are translated into customs declarations.”

• Status: confirmed (evidence score 0.67)
• Findings: 7,863 across 2 signals
• Money at risk: CHF 258.7B (cumulative exposure)
• Evidence: probe_customs_vs_shipment (primary) + probe_customs_duty_reconciliation (supporting) + probe_hs_classification_mismatch (context)

Step 4: Verdicts — why is it happening?

Each confirmed thesis gets a root cause verdict:

Manual HS classification without product master (confidence: 65%)

Root cause: process_failure / manual_classification

“HS codes are assigned manually by different staff without a central product-to-HS mapping table. The same product is declared under different tariff codes across shipments. This inconsistency leads to overpaid duties and compliance risk.”

Recommendation: Build and enforce a product master with standardized HS code assignments. Implement classification validation at declaration entry.

Route capacity mismatch (confidence: 75%)

Root cause: structural / route_capacity_mismatch

“Delivery delays are driven by a mismatch between shipment volumes and route capacity. High-traffic corridors are consistently overloaded, creating cascading delays that exceed SLA windows. The issue is structural — individual carrier performance is within norms, but the routes themselves are saturated.”

Recommendation: Review route allocation and consider splitting high-volume corridors across multiple carriers or adding intermediate hubs. Evaluate whether current SLA targets reflect realistic transit times for peak-volume routes.

Weak carrier contract enforcement (confidence: 75%)

Root cause: structural / weak_contract_enforcement

“Carrier underperformance persists because SLA penalties are either absent or not enforced. Carriers show consistent performance gaps, but no contract consequences are triggered. The commercial relationship prioritizes volume over accountability.”

Recommendation: Introduce tiered SLA penalties with automatic triggers based on rolling 90-day performance windows. Establish quarterly carrier reviews with data-driven scorecards.

Invoice-to-declaration process gap (confidence: 65%)

Root cause: process_failure / invoice_to_declaration_gap

“Customs declared values diverge from shipment values because the commercial invoice and customs declaration are prepared by different teams without a reconciliation step. Some declarations are undervalued (compliance risk), some overvalued (duty overpayment).”

Recommendation: Introduce a pre-submission reconciliation step where customs declarations are automatically compared against the commercial invoice before filing. Flag declarations with value gaps exceeding 5% for manual review.

Step 5: SMEbits — expert knowledge

Four pieces of operational knowledge captured from logistics staff:

smebit_sunday_driving_ban (business_rule)

“European road transport has Sunday and holiday driving bans in Germany, Austria, and Switzerland.”

Regular checkpoint gaps every weekend for road shipments are regulatory, not a tracking failure. This tells the analyst: when probe_delivery_sla_breach flags a weekend delay, check if it’s a DE/AT/CH route before escalating.

smebit_shanghai_customs_48h (business_rule)

“The Shanghai inspection hold — 48h mandatory customs dwell for electronics and pharmaceuticals is regulatory, not a tracking gap.”

Pacific Trade’s apparent dwell time anomalies at Shanghai are by design. This cross-tenant SMEbit prevents false alarms.

smebit_express_no_containers (structural)

“Express Europe operates without containers — all shipments are parcels or pallets.”

Container-related signals return no_data for this tenant by design. A structural SMEbit that prevents misinterpretation.

smebit_cargowise_consolidation_split (system)

“CargoWise splits consolidated shipments into house bills at destination.”

MBL-to-HBL deconsolidation breaks the checkpoint chain by design, not by error. This explains apparent checkpoint gaps when a master bill arrives as one shipment and leaves as many.

Step 6: BitBundle — the curated narrative

bb_three_logistics_data_landscape

“Three Systems, Six Continents, One Chain”

The BitBundle curates 5 SMEbits into a narrative about how InterLogic’s three operations (Express Europe / CargoWise, Global Freight / SAP TM, Pacific Trade / WMS Manhattan) each bring different data characteristics and known limitations.

SMEbit	Note
smebit_sap_tm_timezone_utc	The UTC normalization challenge — SAP stores UTC but carriers report local time, creating apparent time-travel in checkpoints
smebit_shanghai_customs_48h	The Shanghai inspection hold — 48h mandatory customs dwell is regulatory, not a tracking gap
smebit_sunday_driving_ban	The Sunday driving ban — legal truck rest periods create regular checkpoint gaps every weekend
smebit_express_no_containers	The express exception — no containers at all, so container signals return no_data by design
smebit_cargowise_consolidation_split	The consolidation split — MBL-to-HBL deconsolidation breaks the checkpoint chain by design

Step 7: Reports — the deliverable

5 reports (15 PDFs in DE/FR/EN):

Report	What it delivers
report_executive_summary	High-level overview — confirmed theses, total exposure, top recommendations
report_customs_compliance	HS classification analysis, duty discrepancies, compliance risk by product category
report_carrier_scorecard	Carrier SLA performance, incident rates, on-time delivery metrics
report_tracking_integrity	Checkpoint coverage, gap analysis, mean time between scans
report_data_landscape	Data quality overview — completeness, validity, source system characteristics

The full chain

One thread, every instrument:

From a single shipment item (Entity) to a process recommendation (Report) — through every layer of the analytical pyramid. Same engine, different domain, same clarity.

Two packs, one engine

Compare this showcase with From Vine to Verdict — the Millesime winemaking walkthrough. Different industry, different entities, different signals. But the same instrument chain, the same Explorer, the same path from finding to action.

That’s the point of jinflow: declare once, detect everywhere.

Try it yourself

jinflow explore --tenant interlogic.express_europe

Talk to your data — so it speaks to you.