Kia investment in Arrival bears fruit with microfactory van plan

Hyundai and Kia investing £100 million into Arrival back in 2020 looks like money down the drain as the British EV firm continues to sink – to the point that it’s now threatened with delisting from the Nasdaq stock exchange.

Or is it? At the CES tech show in Las Vegas last week, Kia impressed with a line-up of modular electric vans whose promised construction methods looked very familiar.

The Korean company is proposing that its so-called PBV (for Purpose Built Vehicle) commercial range, due from 2025, can be built globally in microfactories in a way very similar to that proposed by Arrival. 

Kia told Autocar that the intellectual property for the vans and their construction methods rests entirely with Kia itself, with nothing taken from Arrival.

“In the past we’ve invested in Arrival but as a financial investor. We’ve developed our own production process,” said Sangdae Kim, Kia’s head of PBVs. 

However, the original investment was made by Hyundai and Kia partly with the goal of exploring the idea of a flexible van platform, and the PBV range is the fruit of this, only this time with the benefit of solid financial backing to (hopefully) see it to fruition.

Kia’s concept for the microfactory is similar to Arrival’s, with a few differences.

Small plants with the capacity to build a few as 1000 vehicles a year (fewer than Arrival’s planned 10,000 annually at its Bicester site) allow localised production of vehicles specifically tailored to needs of nearby buyers.

Car makers have traditionally designed factories to build vast numbers of a handful of models, ideally more than 100,000 per year for each. That way they can maximise efficiency and enrich themselves from the resulting economies of scale.

The downside of this strategy is that these factories are very inflexible when buying trends change, when supply shocks hit or when demand falls. 

Microfactories, on the other hand, are much more flexible. Being smaller, they can also be built more cheaply or in areas with reduced land availability.

For example, Hyundai in November opened a car plant in Singapore with a capacity of just 30,000, utilising some of the microfactory format.

The format itself can be traced to a paper entitled Micro Factory Retailing published back in 2000 by two professors working at the Centre for Automotive Industry Research in Cardiff, Paul Nieuwenhuis and Peter Wells.

Profitability of microfactories is an open question, given that so few have been established so far. LEVC’s taxi factory near Coventry is an example that might one day become profitable if the company can reduce the cost of the parts. 

Kia aims to reduce costs by giving the microfactories much less to do. Instead of stamping and painting parts within the factory walls, the plant will assemble modules shipped from one of its more traditional factories. 

Car companies have long used CKD (completely knocked down) plants that build cars from imported kits to bypass import taxes. However, this is different in that the modules aren’t specific to one model but can be used interchangeably.

The Kia PBV chassis is a skateboard arrangement with an expandable wheelbase that includes the electric motor, battery, inverter and transmission. The body, meanwhile, is a skeleton of interlocking supports that Kia says was inspired by traditional Korean joinery that uses no nails or glue. Plastic bodywork panels that come precoloured are fixed over this.

This way, Kia says it can create a range of vans, starting with the mid-sized, 4.7-metre-long PV5, a rival to the Ford Transit Custom, from the second half of 2025. Also shown at CES was the longer PV7 due in 2027 and the shorter PV3 and PV1, neither of which yet have a starting date.

These are first and foremost vans, but Kia also showed MPV and taxi models, including a robotaxi. The one fixed element is the ‘driver zone’, with the rest customisable.

The microfactories are a longer-term plan. Initially the PBVs will be built at Kia’s Hwaseong plant in South Korea with a capacity of 150,000 a year. A second plant nearby is also planned, according to media reports.

These plants are traditional in terms of their output, but here too the company is experimenting with non-traditional ‘smart’ manufacturing, in which vehicles are built in a series of robot-heavy cells that optimise construction by reducing the time it takes for parts to reach the work station.

Of all the vehicle categories, vans are proving hardest to disrupt. Fleet buyers are often conversative and go with brands they know and trust, which favours Ford and Stellantis in particular. Meanwhile, van makers keep costs low and margins high by building at scale and reusing platforms. 

But the shift to electric gives a new opening to those wanting to elbow out the incumbents, and even established players are testing new concepts. Renault, for example, has said that its first true software-defined vehicle will be its FlexEVan, due in 2026, of which Nissan will also take a version.

In the US, meanwhile, General Motors has rolled out its Brightdrop range of electric vans using its Ultium electric car platform.

Geely-owned LEVC has unveiled its modular Space Oriented Architecture, intended to spawn a range of vans, taxis and MPVs that it hopes will lower the cost from the current high level of the TX range-extender taxi.

The race is on to see which approach works best in the shift to electric vans: the incorporation of batteries in a traditional ICE platform, as favoured by Ford and Stellantis, or Kia’s ground-up redesign utilising smart manufacturing.