Donal Reddington on mass customization, crowdsourcing and digital manufacturing


Car Trouble

There have been a few recent articles from various sources regarding the U.S. auto industry, which provide an interesting basis for a discussion about the role of mass customization in the future of the automotive sector.

In Industry Week, a recent feature by Yorgos Papatheodorou and Michelle Harris, which primarily dealt with the geographical spread of auto plants within the U.S., also notes the changing nature of the auto industry over recent years, and the ever increasing fragementation of the market. They note that:

“This creates a dilemma between the efficiency of mass production and the complex needs of customers. To bridge the two, Toyota pioneered lean manufacturing techniques, which have been adopted by other companies to a greater or lesser extent. However, a cynic might counter that lean assembly, just-in-time and low goods-in-process inventories are all very well, but finished vehicles are still produced for inventory. Typically, 45 to 60 days of inventory is piled up on parking lots, dealerships and in transit. The costs are staggering: billions of dollars in working capital are tied up, while companies run the risk of changing tastes and circumstances, such as gasoline price spikes, which may dramatically alter the demand for vehicles away from forecasts and production. It that case, discounts are needed to “move the metal,” and customers have become addicted to them.”

The article poses the question as to why the manufacturers don’t apply the Dell model, where a car is produced only when there is a firm custom order for it?

Numerous reasons are given for this:

  • Customers don’t want to wait (except for luxury cars), and (especially in the U.S) often even enjoy haggling.
  • Corporate systems are slow at taking, consolidating, and transmitting orders.
  • An extensive system of dealerships is well entrenched and protected by U.S. state and federal laws.
  • Manufacturing really is not flexible enough yet.

The article states that:

“It therefore seems safe to speculate that the auto company of the future will need better margins and a large variety of vehicles with a faster turnover of models. Better margins will require more efficient production and logistics with lower inventories of finished vehicles and more emphasis on vehicle life-cycle services, such as leasing and after-sales service.”

Among a list of implications for the auto industry are the need for:

  • Faster product design;
  • More flexible plants;
  • More commonality of parts and subassemblies.

So it is clear that a much greater level of overall flexibility is required. This will take some years to achieve. In the meantime, can the manufacturers do anything to promote the build-to-order model? The answer may lie in the themes explored in an article by Jerry Marks, writing on the Seeking Alpha website, where he discusses the difficulties currently facing GM and Ford. He mentions that a U.S. car dealer emailed him with some comments, including:

“The ability to adapt to the ever shortening consumer acceptance cycle is not just wishful thinking, but something that will be required out of successful automakers in the future (that it is almost like the fashion industry).”

In response, Marks himself says:

“I agree, the life cycles are shortening to meet the ever fickle and changing demands of the U.S. consumer. BUT, I need to emphasize, throwing more money at the problem (opportunity/challenge whatever you want to call it), is not the solution. I continue to believe automakers need to re-think the notion of the vehicle. I think Scion is really onto something with their almost “mass customization” concept of allowing consumers to accessorize a rather base vehicle in a million different ways. If you ask me, the BASE vehicle needs to become more standard. But the ability to customize and accessorize the vehicle is where the excitement and personalization really come into play. Just my thought, particularly when it comes to the younger generations (gen x and y) at least.”

This comment about making the base of a vehicle more standard is an interesting point – by giving the customer more of a blank canvas, it reduces the percentage of the vehicle which anticipates demand, and increases the percentage that is demand-led.  This concept is very similar as that which was applied by HP in the computer printer market a few years ago, and analysed very effectively in the often-quoted Harvard Business Review article from 1997, ‘Mass Customization at Hewlett Packard – The Power of Postponment’*.  That article described how the power supply units in a printer must vary from country to country due to varying technical standards. Hewlett Packard took two separate approaches to the problem.  In LaserJet printers, a power supply that worked in all countries was built into the product.  For DeskJet printers, however, “the company opted to customize the printers at its local distribution centers rather than at its factories”.   The factories would firstly assemble printers without power supplies, and forward them to distribution centres, where an external power supply unit would be added to the printers, as appropriate, for each country served by the distribution centre.  By not adding the power supply unit until the latest possible time, HP were in a position to match most closely the actual demand in each country at any given point in time.

The distribution centre not only customized the printer for the local markets, but also purchased the materials that differentiated it, such as the power supplies, packaging and materials.

An auto maker who provides a very basic starting point is doing the same thing, except that the auto maker is using postponment for each customer rather than a country.  While the preferences of individual auto customers are far more variable than the electrical standards of a country, the basic principle is the same – the manufacturer delays incurring some of the marginal cost of the product until it is much closer to the time of sale.  In the case of HP and their DeskJet printers, the company found that manufacturing costs were slightly higher than when the factory customized the printers for each market, but the total manufacturing, shipping, and inventory costs dropped by 25%.

In the case of the auto industry, however, there are some significant obstacles that may explain why this strategy has not been widely employed before now.  The first issue is complexity.  While a local distribution centre may easily be able to add cosmetic personalization features to a basic base model, such as wheels, side skirts and other external items, anything more fundamental requires significant investment in plant and labour.  It is unlikely to be economic for local distribution centres to have the role of adding the customers chosen stereo system or fuel intake system.  The second issue is size and weight – car components are a lot bigger and heavier than the power supply or user manual for a desktop printer.  The reduced shipping and inventory costs gained by the likes of HP may not be achieved if the locally purchased components weigh 100-200kg per car.  And lastly, there is the important issue of product liability.  Could a manufacturer be certain that the components purchased locally by a distributor are up to the same standard that it would require of its own suppliers?

These are difficult questions that would require a lot of research on the part of the auto makers.  However, the thing about implementing a demand-led strategy is that those doing the demanding have to want it.  The current customer preference in the U.S. for walking in and buying a finished car from dealer stock means that the build-to-order model cannot easily be applied more widely in the U.S. auto market for now.  If the manufacturers are to reduce their dependency on building for inventory, they must first focus on spreading the popularity of the build-to-order idea to a much greater percentage of their current customers.
U.S. manufacturers were caught on the hop in the 1970′s when sudden oil price hikes made their gas-guzzling cars unpopular with buyers.  During the 1990′s, low oil prices allowed manufacturers to make good profits on big engined SUV’s that had very basic engineering. Investment in other models, like Ford’s Taurus, was cut back.  When oil prices rose again in recent years, the manufacturers found they had made the same mistake as in the 1970′s.

If manufacturers had greater foresight during the 1990′s, they could have at least looked at popularising build-to-order so that they would not fall into the inventory trap if a sudden downturn in sales were to occur.  Even without the political turmoil that has affected the World since the turn of the Millennium, it would have been clear that growing demand would place upward pressure on oil prices, and reduce demand for vehicles with heavy fuel consumption.

The inability or unwillingness to see this coming has put the U.S. auto companies at the bottom of a very large mountain, where they could have started the climb half way up.  The Gen X and Gen Y customers in the U.S. who personalize their Scions now may at a later stage be in the market for a made-to-order Ford or GM product. But by then there may not be any Ford or GM product for them to buy.

If a disruptive technology or manufacturing system were to emerge, these problems would be multiplied even further. This is something I will look at in a separate post soon.

*Edward Feitzinger and Hau L. Lee -Mass Customization at Hewlett Packard – The Power of Postponment, Harvard Business Review Jan-Feb 1997.

Share this item on these services:
  • del.icio.us
  • digg
  • Furl
  • NewsVine
  • RawSugar
  • YahooMyWeb

Leave a Reply

You must be logged in to post a comment.