ALLWIN AIRBusiness Model
AI Seat Recommendation & Technology

Not a seat map —
the right seat for the situation, first

Show hundreds of empty seats and travelers hesitate. ALLWIN AIR reads the traveler's situation from the booking and offers the seats that fit it first.

Contact partnershipsPartnership info

Unsold seats are the market

Most aircraft seats are still assigned for free. Only a few rows such as the front and exit rows sell; the rest mostly stay unsold. Give each seat a reason that matters to the traveler, and these seats become new revenue.

Bar chart of paid vs free seat assignment by row on Jeju Air (rows 1–32). Only a few rows such as rows 1–3 (20%, 19%, 13%), row 12 (17%) and row 23 (19%) have high paid rates; most rows are 0–10%. Overlay: 'Opportunity'; top right: 'Potential market KRW 100T'.
Most seats are still assigned for free. Unsold seats are the market opportunity.
Diagram 'A seat is a human desire'. Traveler needs grouped as 01 Tussling, 02 Bonding & 03 Caring, 04 Toying, 05 Hunting, with eight seat types on seat maps: 1 Fastest (front, first off), 2 Baby Caring, 3 Guarantee Group Privacy (group side by side), 4 Romantic, 5 Talking, 6 Viewing (window), 7 Moving (aisle), 8 Individual Privacy. Bottom: 'There is no such thing as a bad seat.' 'Give every seat a meaning, and every seat becomes valuable.'
Give every seat a meaning and every seat becomes valuable — the starting point of ALLWIN AIR seat recommendation.

We read the situation from the booking (PNR)

Seat selection rates differ widely by who is in the booking — alone, a couple, with children, with seniors, a group. ALLWIN AIR uses this data to decide which seats to show to whom first.

'Which travelers are most likely to pick a seat?' — horizontal bar chart of seat selection rate by PNR type: Other 32.5%, 2 seniors M+F 22.2%, adult+child 12.5%, 2 adults M+F 11.1%, 1 adult 6.5%, adult+senior 2.8%, adult+infant 1.9%, adult+child+senior 1.6%, 2 adults 2F 1.6%, adult+infant+child 1.3%, 3 adults M+F 1.0%, 4+ adults M=F 1.0%, 4+ adults men only 0.8%, 1 senior 0.6%, child only 0.6%, and others.
Seat selection rates differ widely by party composition in the booking (PNR). We recommend based on this data.

How we recommend

We show the chance that the party will sit apart without selecting seats, and immediately offer seats that fit.

Seat recommendation example 1: 'Traveling with a child? How about a roomy front seat to care for your child?' Two adults and a baby icon; chance of sitting apart without seat selection: 99.5%. Below, a seat map with front and exit seats and prices per row.
Traveling with a child — we offer roomy front seats first. Chance of sitting apart without selection: 99.5%.
Seat recommendation example 2: 'Traveling as a couple?' Two adults; chance of sitting apart without seat selection: 88%. Below, a seat map with recommended side-by-side seats.
A couple — seats side by side. Chance of sitting apart: 88%.
Seat recommendation example 3: 'Would you like to sit with your group?' Three adults; chance of sitting apart without seat selection: 85%. Below, a seat map with three seats together.
A group of three — together in one row. Chance of sitting apart: 85%.
Seat recommendation example 4: 'How about seats that make talking easy?' Four adults; chance of sitting apart without seat selection: 99%. Below, a seat map recommending facing seats based on space, distance, cost and view.
Four travelers — a layout made for conversation. Chance of sitting apart: 99%.

Why conversion goes up

'Conversion Winner?' balance illustration. Left: 'Airline seat SYSTEM — per-passenger, plain display' (seat map and prices only). Right: 'ALLWIN seat SYSTEM — party-level, AI-predicted display' (seats grouped as front/exit, zone A, zone B, fixed-recline zone B). The balance tips toward ALLWIN. App screenshots are original Korean screens; personal data is masked.
A plain seat map versus party-level predicted recommendations — recommendation wins on conversion.
AspectTypical airline seat selectionALLWIN AIR seat selection
UnitEach passengerThe whole party, from the booking (PNR)
DisplayPlain seat mapRecommended seats grouped by AI prediction
ExperienceSearch and compare empty seatsPick a recommendation and pay

Patents

PNR-based patented engine diagram 'C.R. (Conversion Rate) Engine'. Aircraft seats (Seat) on the left and duty-free goods (Dutyfree) on the right linked through PNR (Passenger Name Record) companion data. Seats: global potential seat selection market (2019 tickets USD 899B × 10% ≈ USD 90B). Duty-free: Korean duty-free market (KRW 24 trillion in 2019). AGS engine (one-stop buying, even less-preferred seats): aircraft seat pre-purchase method & device (10-2021-0035677), item recommendation method & device (10-2020-0069932), auction method & server (1115802 etc., for seat upgrades). DGS engine (one-stop duty-free for companions within allowance): duty-free pre-order method & device (10-2021-0023742), duty-free display control method & device (10-2021-0057065), item recommendation method & device (10-2020-0069932).
A conversion-rate (C.R.) engine that recommends seats and duty-free around the booking's companion data — with filed and registered patents.
Rational pricing engine for seat upgrades. Left: 'Auction method & server' patents registered in 12 countries including the USA, Japan, Korea, Indonesia, Israel and Singapore (Group Auction, Dynamic Pricing). Right: Group Auction — customers bid between MIN and MAX, top bids win up to the set quantity, all winners pay the lowest winning price, and no one pays more than their bid. Dynamic Pricing — real-time pricing between MIN and MAX like a stock market; high demand raises the price (higher airline margin), low demand lowers it (higher conversion); customers buy instantly without waiting.
Pricing patents for seat upgrades (group auction and dynamic pricing) registered in 12 countries.
Alpha Seat for Airline → Musical → Arena: the same seat recommendation engine applied to aircraft cabins, theater seating and ballpark seating charts.
The recommendation engine extends beyond aircraft to any seated booking, such as theaters and arenas.

Frequently asked questions

How does AI seat recommendation work?
It analyzes the booking's (PNR) party composition and itinerary and shows recommended seats first, considering whether the party can sit together and seat attributes such as front, exit, window and aisle. Travelers pick a recommendation or adjust it and pay.
Is it only for airline seats?
It is designed to work for any seated booking, including theaters and arenas.
Is the technology patented?
Patents have been filed for PNR-based seat pre-purchase, item recommendation and duty-free pre-order, and auction/pricing patents for seat upgrades are registered in 12 countries.

Start integrating with ALLWIN AIR

Airlines and ancillary suppliers, as well as travel agencies, OTAs and travel platforms, can all start with a single integration. We agree on the integration model and schedule to fit each partner.

Partnership e-mail (recommended from overseas)partnership@allwin.co.kr
Main phone (Korea, weekdays 09:00–18:00 KST)+82 1522-2284