Point of viewAirline industry trends

Virtual interlining: selling the itineraries the network can't

Budget travellers will take more stops for a lower fare, but code-share itineraries price them out. Treating the global flight network as a graph — with real connection rules built in — opens itineraries traditional channels never show.

6 min read By · Point of view
392
virtual route combinations from 22 airlines in an illustrative demo search1

Key takeaways

  • A segment of travellers prioritises price over comfort and will accept multi-hop journeys that code-share routes rarely offer.
  • Modelling airports as nodes and flights as edges lets an engine search combinations across carriers that have no interline agreement1.
  • The hard part is feasibility: minimum connection times, terminal changes, baggage and transit-visa rules must prune impossible routes before ranking1.
  • On one New Delhi–San Francisco example, the engine analysed 22 airlines and 392 virtual combinations, and found an option 18% cheaper and 10 hours faster than a leading OTA's cheapest — an illustrative demo search, not a measured outcome1.

Most booking channels show the itineraries that airlines have agreed to sell together. That leaves out a large space of journeys that combine low-cost and full-service carriers with no agreement between them. For travellers who put price first, that space is where the value is.

The network as a graph

The approach in our delivered work treats every airport as a node and every flight as an edge carrying fare, times, carrier type, baggage rules and connection constraints. Content from GDS, NDC and direct LCC APIs plugs into the same graph, so adding a carrier means adding edges rather than redesigning the system1.

Feasibility before optimisation

A cheap route that misses its connection is not cheap. The engine checks minimum connection times per airport and carrier, terminal changes, immigration time for international transfers, self-connect baggage requirements and transit-visa eligibility by passport, and drops infeasible edges before any ranking happens1. Only then does a multi-criteria optimisation rank the fastest, cheapest and best-value journeys.

Exhibit 1

New Delhi to San Francisco: leading OTA vs our engine

Example search from our delivered OTA-marketplace work

SourceOptionTravel timeCost (USD)
Leading OTACheapest35h 30m856.70
Our engineCheapest24h 20m702.44
Leading OTAFastest15h 30m1,405.72
Our engineFastest15h 30m1,044.54

Note: 22 airlines analysed, 392 virtual combinations.

Source: DaasLabs, “Transforming aviation data into actionable intelligence: travel-technology case study” (2026)

For OTAs, the prize is a traveller segment that traditional channels underserve. For airlines, the same graph is a planning tool: it shows which partners and connections would open demand the network cannot capture today.

For executives

What this means for your bank

  1. Normalise GDS, NDC and LCC content into one graph before building ranking logic.
  2. Encode MCT, terminal, baggage and visa rules as hard constraints, not after-the-fact warnings.
  3. Rank on more than price: travel time, layover safety and connection risk belong in the score.
  4. Use the same graph for partner and network-planning decisions.
Put it to work

How DaasLabs can help

See the virtual-interlining case study, including the DEL–SFO comparison.

Learn more

Build interlining and distribution data products with our Commercial, Distribution & Revenue service.

Learn more

Explore the Data Fabric Framework that connects GDS, NDC and direct content.

Explore the framework

Sources

  1. 1
  2. 2

Figures are drawn from the cited public sources. Opinions labelled “DaasLabs point of view” are our own.

Stay informed

Get new airline insights in your inbox

New perspectives on AI, data and transformation in airline — a few times a month. Browse all insights.