Hub-and-Spoke vs Point-to-Point in Southeast Asia: Three Network Models Compared

The way an airline structures its network is the most consequential strategic decision it makes. Fleet selection, slot strategy, hub investment, route launch timing, alliance membership, even the way the airline brands itself to customers all flow downstream from one choice: do passengers connect through a central hub, or do they fly direct between origin and destination?

In Southeast Asia, this choice is more interesting than in most regions. The geography is fragmented across thousands of islands, mountain ranges, and political boundaries; the wealth distribution is uneven, with high-income city pairs sitting alongside low-density routes that only an LCC can sustain; and the regulatory environment includes both deeply liberalised channels (ASEAN MAAS) and traditional bilateral relationships. Every ASEAN carrier has had to choose a network architecture suited to this environment, and the three dominant models that have emerged each have clear strategic logic, distinct fleet implications, and predictable failure modes.

This article walks through the three models, uses Singapore Airlines, AirAsia Group, and Garuda Indonesia as case studies of each, and concludes with the trade-offs and decision factors that determine which model fits which carrier.

1. The two pure models, and why they conflict

The hub-and-spoke model concentrates flights at one or a small number of central airports, with passengers from secondary cities connecting through the hub to reach their final destination. The point-to-point model flies passengers directly from origin to destination, without requiring connections.

The conflict between the two is fundamental, not stylistic. Hub-and-spoke uses fewer aircraft and crew to serve more origin-destination pairs, because each “spoke” city only needs flights to the hub rather than to every other destination. A hub with 50 spokes generates roughly 1,225 indirect origin-destination pairs (the combinations of 50 cities taken two at a time, all connecting via the hub), against only 50 actual routes. The economics are about combinatorial leverage.

Point-to-point operates without this leverage. Each city pair requires its own flight, with its own aircraft, crew, slot, and ground handling. To serve the same 1,225 origin-destination pairs as direct flights would require 1,225 routes, which is operationally impossible for any single carrier. Point-to-point therefore necessarily covers far fewer total origin-destination pairs, but offers each of those pairs a faster journey (no connection time, no risk of misconnect, no terminal walk).

The trade-off is captured in a single sentence: hub-and-spoke maximises network reach with minimum aircraft; point-to-point maximises customer convenience with maximum aircraft per market served. Which trade-off matters more depends on geography, demand density, and customer mix.

2. The three models in practice

Pure hub-and-spoke and pure point-to-point both exist, but most ASEAN carriers actually operate one of three model variants:

Model A: Hub-dominant. A single primary hub absorbs nearly all connecting traffic. Spokes feed inbound, long-haul departures consolidate, returning passengers re-connect to spokes. The fleet skews heavily toward widebody equipment for long-haul departures, with narrowbody backup for short spoke feeds.

Model B: Distributed point-to-point. Many origin cities, many destination cities, direct routes between them where demand supports. The fleet is overwhelmingly narrowbody, with a small widebody overlay for the longest routes if they fit the operator’s range capability.

Model C: Hybrid (domestic hub + international point-to-point, or international hub + domestic point-to-point). A single hub serves international connectivity, while domestic routes are flown point-to-point. Or the inverse: a single hub serves domestic connectivity to international gateways, while long-haul routes are flown point-to-point. Most national flag carriers in middle-tier economies operate some version of the hybrid model.

Each of the three models has a real-world exemplar in Southeast Asia. The next three sections work through one each.

3. Case study: Singapore Airlines — the canonical hub model

Singapore Airlines operates the most explicit hub-and-spoke model in Southeast Asia. The mainline passenger fleet as of mid-2026 consists of approximately 144-149 aircraft (sources vary on exact counts in transition periods), with the structural composition heavily skewed widebody:

• A350-900 (multiple variants including medium-haul, long-haul, and ULR): around 65 aircraft total
• Boeing 787-10: 28 aircraft
• Boeing 777-300ER: 21-22 aircraft
• Airbus A380-800: 12 aircraft
• Boeing 737-8 MAX (narrowbody): rising toward a planned 29 aircraft

Widebody aircraft account for over 85% of the fleet. The narrowbody 737-8 fleet, mostly inherited from the SilkAir merger, exists to feed regional spokes that do not justify a widebody. The strategic centre of gravity sits firmly on long-haul widebody operations from Singapore Changi.

The network architecture reflects this fleet composition. Singapore Airlines does not fly between two foreign cities; it flies from foreign cities to Singapore, and from Singapore to other foreign cities. A passenger flying from Jakarta to Frankfurt connects in Singapore. A passenger flying from Mumbai to Sydney connects in Singapore. The hub absorbs the connecting flow, the widebody fleet executes the long-haul departures, and the spoke aircraft feed inbound.

The strategic rationale is geography combined with market positioning. Singapore is a small city-state with limited origin-destination demand of its own, but it sits at a useful point between South Asia, Northeast Asia, Europe, and Australia. By positioning Changi as the connection point and investing in premium product, Singapore Airlines captures a share of premium long-haul connecting traffic that geographically larger competitors cannot match purely from their home markets.

The recent ultra-long-range expansion (the A350-900ULR fleet and the planned 10 new ULR routes for 2026) extends this hub logic outward, reaching ever-more-distant city pairs that bypass intermediate connections. The ULR strategy is not a departure from hub-and-spoke; it is hub-and-spoke at maximum range, allowing the hub to capture demand from cities that previously would have required two connections.

The model’s vulnerabilities are also clear. Singapore Airlines is exceptionally exposed to anything that disrupts the Changi hub: a major airport closure, a regional infectious disease event, or a sustained Singaporean economic downturn would compromise the entire network. The 2020-2022 COVID period exposed this exposure brutally, and the carrier’s recovery has hinged on rebuilding hub connectivity rather than diversifying away from it.

4. Case study: AirAsia Group — the point-to-point exemplar

AirAsia operates the most explicit point-to-point model in the region. As of mid-2026, the group fleet (including AirAsia Berhad in Malaysia, Thai AirAsia, Indonesia AirAsia, AirAsia Cambodia, and AirAsia Philippines) consists of approximately 239 aircraft, planned to reduce to around 229 by mid-2026 as the group returns 14 older aircraft to lessors. The fleet is almost entirely Airbus A320 family narrowbody:

• A320ceo: the largest single subtype, being progressively retired
• A320neo: the workhorse for current operations
• A321neo: the higher-capacity variant for high-demand routes
• A321LR (first delivered April 30, 2026): for medium-haul reach
• A321XLR (on order): for extended single-aisle range

The A220-300 order of 150 aircraft announced in May 2026, the single largest A220 order in history, is intended to replace older A320s starting in 2028. AirAsia Group also retains a small A330 fleet through AirAsia X for longer-haul operations.

The network architecture matches the fleet. AirAsia’s five national subsidiaries each operate their own dense web of point-to-point routes, with hubs at Kuala Lumpur (KUL), Don Mueang (DMK), Cengkareng (CGK), Sihanoukville (KOS), and Manila (MNL), but these are origination hubs rather than connection hubs. A passenger flying Bangkok-Bali on AirAsia takes a direct flight; they are not expected to connect through Kuala Lumpur. The hub-and-spoke logic of consolidating long-haul departures simply does not apply, because AirAsia’s product is short-to-medium-haul direct flights at low fares.

The strategic rationale is the market opportunity created by Southeast Asia’s rising middle class. As regional incomes grew through the 2000s and 2010s, demand emerged for direct travel between specific city pairs that flag carriers could not serve economically. AirAsia’s low-cost model (single aircraft type, no premium product, secondary airports where possible, fast turnarounds, ancillary revenue) made these previously unviable city pairs profitable. The model scales by adding more aircraft and more city pairs, not by deepening any single connection point.

The model’s vulnerabilities include sensitivity to fuel prices (a high-utilisation, low-margin model has limited buffer when fuel spikes), competitive vulnerability when premium carriers add capacity at low fares, and concentration risk within Southeast Asia (the model has not exported successfully to other regions). AirAsia X’s various long-haul experiments illustrate the difficulty of stretching point-to-point logic into long-haul markets where the unit economics are fundamentally different.

5. Case study: Garuda Indonesia — the hybrid model

Garuda Indonesia operates a more complex model, partly by design and partly by historical accident. The carrier serves approximately 96 destinations across 12 countries (72 domestic and 24 international), with around 500 daily departures from its primary hubs at Jakarta (CGK) and Denpasar (DPS), plus secondary hubs at Makassar and Medan. The fleet of approximately 140 aircraft includes a mix of widebody (Boeing 777-300ER, Airbus A330) and narrowbody (Boeing 737-800, ATR 72) types.

The architecture is structurally hybrid because Indonesia’s geography demands it. With over 17,000 islands across 5,000 km of east-west extent, the country cannot be served as a single hub-and-spoke network. Jakarta serves as the primary hub for connecting international flights and for connecting east-west domestic flow, but secondary regional hubs handle local distribution that would be uneconomic to route through CGK.

Internationally, Garuda operates a more hub-dominant pattern: long-haul widebody routes (Tokyo, Amsterdam, Sydney, Doha when running) depart from Jakarta and Denpasar, with passengers from secondary Indonesian cities connecting through these hubs. Domestically, the pattern is more point-to-point: passengers flying Surabaya-Makassar do not connect through Jakarta, but take direct flights.

A further wrinkle: Garuda has progressively transferred lower-yield domestic routes to its budget subsidiary Citilink, which operates its own point-to-point network with all-economy A320 aircraft. The Jakarta-Bengkulu route, for example, transitioned from Garuda to Citilink on April 1, 2026, with Garuda concluding its B737-800 service on March 28 and Citilink stepping in with daily A320 operation. This pattern of group-level network optimisation, where the full-service brand handles higher-yield and primary trunk routes while the LCC subsidiary handles secondary and price-sensitive routes, is common across hybrid-model carriers.

The strategic rationale is the dual mandate that hybrid carriers typically face: they are national flag carriers expected to provide domestic connectivity to all parts of the country (including unprofitable secondary routes), and they are commercial businesses expected to compete on international routes against pure hub operators (SQ, CX) and pure point-to-point operators (LCCs). The hybrid model is the only architecture that addresses both mandates, even if it does neither perfectly.

The model’s vulnerabilities are the costs of complexity. Operating two fleet families (widebody and narrowbody), maintaining two cabin product standards, scheduling against the very different fleet utilisation patterns of long-haul and short-haul, and managing the brand boundary with the LCC subsidiary all impose overhead that pure operators avoid.

6. The fleet implications

The three models impose very different fleet compositions, and a carrier that chooses one model has effectively chosen the bulk of its fleet plan with it.

For hub-dominant operations, widebody concentration is the structural reality. Long-haul departures need range, capacity, and premium product, all of which widebodies provide. The narrowbody fleet exists only to feed the hub; it is not the strategic centre of the operation. Singapore Airlines’ 85%+ widebody share is at the extreme end, but any hub operator targeting long-haul international connectivity will look broadly similar.

For point-to-point operations, narrowbody is the entire story. The A320 family (and increasingly the A220 in next-generation fleets) provides the right combination of capacity, range, and unit economics for the medium-haul direct flights that constitute the model’s core product. AirAsia Group’s near-total A320 family concentration is the natural endpoint of this logic; any operator running the point-to-point model at scale converges on a similar fleet profile.

For hybrid operations, mixed fleets are unavoidable. Some routes need widebody capacity; others need narrowbody efficiency. Carriers that try to consolidate onto a single type either lose markets they cannot serve (widebody-only) or lose competitiveness on long-haul (narrowbody-only). Garuda’s three-type widebody fleet (777-300ER, A330) plus its narrowbody 737 fleet plus the ATR fleet for very short routes is the typical hybrid carrier signature.

The implication for fleet planners is that the network model choice should precede the fleet decision, not follow it. A carrier that orders widebodies before deciding whether it will operate hub-dominant or hybrid is committing to a model implicitly. Better to make the strategic choice first, then build the fleet to fit.

7. The geography that shapes the choice

A persistent question for new carriers entering Southeast Asian markets is which model fits best. The answer is rarely purely strategic; it is heavily constrained by geography and existing market position.

Geographic factors that push toward hub-and-spoke include:

• Home country is small (Singapore, Brunei)
• Home market origin-destination demand is limited relative to connecting potential
• The home airport is geographically well-positioned between major external markets
• The carrier targets premium long-haul international as its primary segment

Geographic factors that push toward point-to-point include:

• Large home country with multiple sizeable city pairs (Indonesia, Philippines)
• Strong regional market with high direct-demand density
• Low average yield, requiring high frequency and direct service
• The carrier targets cost leadership and middle-class growth markets

Hybrid models tend to emerge in countries with:

• Geography that demands domestic distribution (archipelagos, mountains, large landmass)
• National flag carrier status requiring service to unprofitable secondary destinations
• Sufficient international ambition to operate long-haul widebodies
• Government policy or alliance commitments that constrain pure cost optimisation

For mid-sized regional carriers operating in landlocked or single-island countries, with limited home-market origin-destination demand but constrained access to major hub potential, the choice is often the most difficult. Some adopt micro-hub strategies (concentrating flow through one airport for operational efficiency without claiming hub-and-spoke market ambition); others adopt point-to-point strategies focused on a few high-yield routes; some experiment with niche connecting opportunities that larger hubs do not prioritise. None of these intermediate positions are easy to execute, and many regional carriers churn through model choices over a decade as they try to find a sustainable footing.

8. What is changing, and why

Several trends are reshaping the choice between models in 2026:

Single-aisle range extension changes the economics. The A321LR (delivered to AirAsia in April 2026) and the upcoming A321XLR allow narrowbody point-to-point operations on routes that previously required widebody hub-and-spoke architecture. A 7-hour A321LR flight from Bangkok to a Middle Eastern destination is operationally and economically viable in a way that simply did not exist a decade ago. Point-to-point carriers are gaining reach.

Ultra-long-range widebody capability extends hub reach. Singapore Airlines’ A350-900ULR operations, and the planned introduction of the A350-1000ULR by Qantas, push hub-to-distant-spoke direct service further than ever. Long-haul carriers are gaining new reach at the other end of the spectrum.

Group-level network optimisation. Carriers operating multi-brand structures (Garuda + Citilink, Singapore Airlines + Scoot, AirAsia Group’s five national subsidiaries) increasingly route specific city pairs to the brand best suited to serve them, rather than relying on a single uniform network model across all routes. This is the practical evolution of hybrid thinking, and it is gradually becoming the norm rather than the exception.

Regulatory liberalisation through ASEAN MAAS continues to remove constraints on intra-regional point-to-point operations. As multilateral fifth-freedom rights become available between more city pairs, the structural advantage of operating from a national hub erodes for routes that can now be flown by any qualifying ASEAN carrier directly.

9. Three rules for the network strategist

For carriers reviewing their network architecture, three working rules:

Rule 1: Choose the model before choosing the fleet. Every fleet decision is downstream from the network choice. Buying widebody capacity without a hub strategy is buying capacity with no clear deployment plan. Committing to narrowbody-only without thinking through long-haul implications closes options that may matter later.

Rule 2: Geography wins arguments with strategy. A carrier based in a country with strong domestic origin-destination demand will find it expensive and difficult to become a pure hub operator. A carrier based in a small city-state will find it difficult to operate large-scale point-to-point. Most attempted model shifts that fight geography fail.

Rule 3: Hybrid is harder than either pure model. The two pure models, hub-and-spoke and point-to-point, each have well-understood economic logic and fleet implications. The hybrid model has to fight both pure-model competitors with neither the hub depth nor the narrowbody cost base they enjoy. Carriers that choose hybrid should be honest about the strategic complexity they are taking on.

Closing

The network architecture choice is the foundation that every other decision in airline planning sits on. The fleet is downstream of it. The slot strategy is downstream of it. The alliance partnerships are downstream of it. The customer-facing brand positioning is downstream of it. Getting this choice right early in a carrier’s lifecycle saves substantial cost and strategic time later; getting it wrong locks in expensive mistakes that take decades to unwind.

Southeast Asia has been a remarkable laboratory for testing all three models at scale. Singapore Airlines has shown what a pure hub operator can achieve at the premium end. AirAsia Group has shown how far point-to-point can scale when the demand environment supports it. Garuda Indonesia, with all its complexity, has shown that hybrid models can survive even under difficult market conditions, provided the operator manages the inherent complexity carefully. The next decade, with its single-aisle range extensions, its ULR widebody expansions, and its progressing regulatory liberalisation, will produce variations on these models that none of the current operators have yet shown. The fundamentals, however, will remain the same: choose the model, build the fleet, build the network, defend the position.

---

Fleet sizes, route counts, and operational data cited in this article reflect publicly reported information for the named carriers as of mid-2026 and are subject to change. Exact aircraft counts can vary between sources due to ongoing fleet movements, retirements, and deliveries. Strategic interpretations of each carrier’s network architecture are this author’s analysis based on observable operational patterns and are not statements from the carriers themselves.