Behind the scenes at Toyota’s R&D center – Part II

26 Jul 2010 Sandeep Mehta
As we discussed yesterday, Toyota is out trying to mend its broken reputation.  When the light of public scrutiny is shining on a company, it is not good to have shallow marketing campaigns…  Unfortunately, that is what Toyota did with is Star Safety System.  Some actions such as starting field quality offices may have an indeterminate impact.  
On the other hand, as part of this process, they have allowed valuable and unprecedented access to a few journalist to their R&D organization.  This access included detailed briefings about R&D organizations/processes, what might be wrong with them and what they plan to do about them (Deep-Dive: Behind the scenes at Toyota’s R&D center, Part 1 — Autoblog, Deep-Dive: Behind the scenes at Toyota’s R&D center, Part Two — Autoblog).  We discussed existing organization and processes yesterday.  These R&D briefings do actually have some information about concrete steps that Toyota plans to take.  Lets dig into them and see what we can learn from them:

To start Toyota believes the root cause of quality problems is rapid growth and inability of the organization / culture / processes / training to keep up (from Autoblog):

 Takeshi Uchiyamada, Toyota’s executive vice-president for research and development, acknowledged during a group interview that overly aggressive growth over the past decade had contributed to the current problems. Branching into too many new market segments too quickly stretched Toyota’s resources, making it difficult to develop young engineers and technicians.
The assertion is that the lack of oversight is what led to problems – even some simple problems like floor mats sticking under pedals are blamed on inexperience:

The excessively lean organization at Toyota has led to younger staff not getting the necessary oversight to help them learn the nuances in engineering. Engineering is about much more than hard numbers and quantitative analysis – good engineers learn to think outside the box, examining ways their products could be used or misused in unexpected ways. In Toyota’s case, a prime example was the use of all-weather floor-mats. When the mats were developed, they were not intended to be used in conjunction with standard carpet floor-mats – and yet, that’s exactly what happened, leading to a spate of issues with mats being jammed under accelerator and brake pedals.

Based on this root cause analysis, Toyota is proposing several solutions.  From Edumunds:

Some of what Toyota is doing represents “enhancements” of what it has always done, but other steps, such consolidating groups whose areas of responsibility overlap considerably), are more than that.

One proposed solutions is to do more training:

Uchiyamada says that Toyota will be doing “more teaching of younger staff” in ways to examine issues and find innovative solutions.

The reason for the lack of training is thought to be an excessively lean or excessively flat organization:

Toyota now believes that its product development organization has become too flat over the years, with group managers having too many team members reporting to them. While many organizations have been trying to take out layers of management in recent years to improve organizational efficiency and lower costs, this strategy can be taken too far. One of the things that managers need to do is educate and develop the staff reporting to them.
Training (Via Autoblog)

Clearly, more training is generally better.  However, adding a layer of management is not necessarily the best approach to more training.  What Toyota needs is more agility to deal with evolving technologies and changing market conditions.  One of the problems Toyota and most other companies face is faster pace of product development because of increased competitive pressures.  In many cases, extra layers of management actually slow down organizational learning because they want to do things the old way.  More about this is below.  Even more importantly, Toyota lean culture has developed over decades.  How will adding extra layers of bureaucracy change the culture?  Will it take away from the good parts of lean as well as bad?

Another solution proposed is to drive quality innovation through a separate organization and allocation of extra hours:

Each Toyota employee must be “doing our work better,” and that the new Design Quality Innovation Division would be an important factor. He acknowledged the division and its work aren’t necessarily “eye-catching” from a PR standpoint, but are critical.

The four-week extension of lead time invoked by the Design Quality Innovation Division is a very important move, and represents a significant shift on Toyota’s part. This additional time period will not be used for new testing and evaluation per se, but rather gives the division a chance to play devil’s advocate, and look for potential issues from a broad perspective.

Clearly, adding extra time to quality control will hopefully find more flaws.  However, it is extremely important to measure the value of delivered by this additional time.  The devil is in the details.  Putting in extra time to check will not necessarily improve quality unless these inspections are tied into overall R&D process.  Even more importantly, setting up a new division is fraught with dangers – how will this division work with existing quality division, who will be responsible for what?  Who will have ultimate authority?
Another solution proposed is to slow down development and add schedule for additional quality control:

I am not sure how easy it will be implement this… Are customer requirements not driving the design to start with? Why is not customer’s viewpoint part of the overall development through marketing input and reviews?  Why is there a need for a separate evaluation?  What if this review finds problems with a supplier part?  Do they stop the entire development cycle and wait?

To that end, Toyota want to work closely with suppliers (co-design?):

Toyota will coordinate and cooperate more closely with suppliers. Rather than just letting them design to a specification, they will work more closely with Toyota engineers so that Toyota is familiar with their thought and evaluation processes. However, he also said that Toyota will likely be bringing more of the design and development work in house in the future.

    This is clearly a great idea.  However, the quality control and management bureaucracy steps added above might actually impede collaboration with suppliers.  How will Toyota manage these contradictions between processes?

    Finally, I remain convinced that the root causes is actually increased complexity.

    Regardless of whether you’re talking about the most basic transportation in the world (think: Tata Nano) or an advanced hybrid or electric vehicle, it would be impossible to meet the often contradictory requirements of customers and regulators without electronics and software. As the capabilities of electronic systems have increased, so, too, have the complexity of the interactions in these systems. Developing robust electronic control systems requires endless testing at every level, from the earliest software-in-the-loop simulation to full vehicle-in-the-loop evaluation.

    Here is a chart showing all the systems and subsystems linked together through the Engine Control Module (ECM).

    The engine management system alone consists of some 800,000 lines of “C” code split into 1,600 functional modules. Like most manufacturers today, Toyota is using software development tools like Matlab and Simulink to model functions and test them before ever generating a single line of code. Just as simulation is used for developing crash structures, mathematical models of the vehicle and powertrain components are used to check out the software before prototype electronics are produced.

    This complexity will only increase with time.  The pressures to deliver products more quickly will also increase – not reduce.  I am not sure adding time to product development will solve anything.  They really need to go back to the drawing board and identify new processes that will help them be more agile – not add more quality control…

    Leave a Reply