Joe Justice on Rapid and Agile Industrial Development at Wikispeed

From P2P Foundation
Jump to navigation Jump to search

= using agile software development, modular design, and rapid prototyping, the WikiSpeed car development team, developing a 100 MGP car for the Automotive X-Prize, has achieved an extraordinary compression of development time

TedXRanier video presentation of Joe Justice at http://www.youtube.com/watch?v=MkmyKmQppq8&feature=youtu.be ; longer detailed explanation at http://www.youtube.com/watch?v=pjempylJy1w

This is a must see video!


Description

1.

"Joe Justice presents at TEDxRanier how Team WIKISPEED has accomplished amazing results in producing a 100 Mile per Gallon (MPG) car using processes borrowed from the software world; Agile, Lean, Scrum and Extreme Programing. The result is that WIKISPEED now sells an affordable car that achieves 100 MPG that is is fast, fun and beautiful while also achieving a five star crash test rating equivalent."


2.

Charlie Rudd:

"Here are some highlights of what Joe and the others from Wikispeed have accomplished:

  • Designed and manufactured a 4-passenger street-legal car that gets 100 mpg


Most of us would agree that is quite an achievement.

There is more:

  • The car was constructed using off-the-shelf parts


This means, among others things, that the car is easily serviceable using existing maintenance infrastructure

  • The car is entirely modular in design


All sub-systems are essentially snap-out/snap-in, making replacement of engine, brakes, suspension, etc. a process that takes just a few minutes. BTW, this is not simply replacing like-for-like but also to switch to a new technology (e.g. change to a new engine).

  • They innovated a new process for carbon-fiber body construction that costs 1/360th the traditional process


With this process, they were able to switch to a new body type for less than $1000.

  • You can pre-order cars now for less than $29,000


This is not just a one-off prototype. Currently they are manufacturing one car per week (yes, that’s the low volume manufacturing retail price). They are targeting a future price of under $20,000.


Last but not least, all this has been accomplished:

  • With no capital investment


Although they do solicit donations through PayPal on their Web site.

  • No paid employees

Everything is done by volunteers." (http://www.solutionsiq.com/the-agile-ceo/bid/51480/Agile-Innovation-or-How-to-Design-and-Build-a-100-MPG-Road-Car-in-3-Months)


Interview

Summary:

"Here are some highlights from Simone’s interview:


Wikispeed builds ultra efficient cars and we do this with seven days development cycles using agile methodologies. Those methodologies include several aspects, for example about managing distributed teams – like with SCRUM – or methods to ensure an high quality bar and focused work – much like Extreme programming and Test Driven Development as part of XP – that we reworked for the manufacturing process labeling it Extreme Manufacturing.

Wikispeed is missionized to rapidly solve problems for social good. We don’t just make cars. I recently gave a talk on methods for vaccine distribution to help eradicate Polio: we worked with a group that develops low cost medical centers and communites for this and we’ve done significant work with them.

We do this with OSE (Open Source Ecology). OSE gives us the platform to share ours designs, our budgets, our build practices and our maintenance videos with the world so that people can build their own ultra low cost commuter cars.

Instead of us going into a traditional employer that keeps a larger percent of our work…everyone will have an equitable share that is proportional to the amount of value he contributes to the world. We’re not there yet but we’re moving closer: open source hardware movements are making this more possible than ever before.

Will traditional manufacturing disappear and will people be involved themselves directly in the manufacturing process of their own goods? There’s a trend towards localization and communityavailability of manufacturing that we haven’t seen since the trade guilds days – particularly in Europe – and with these available pieces of manufacturing like theGlobal Village Construction Set (GVCS), the hackerspaces and fablabs maybe someone in your community will build your next bread owen or window garden, your next cat feeder or your next pair of pants…I do see this as a fairly near opportunity that will co-exist with traditional manufacturing and eventually be growing in market share.

Obviously we can still not print core i7 miniaturized size circuitry but we are getting closer to this and by thanks to the open source community we’ll be able to iterate and improve so I do see us, able to produce our next quad core CPU and I do see us able to reduce and reduce and reduce the amount of energy needed to do this.

We need absolutely to develop more energy and resource efficient ways of production and maybe local production and local manufacturing coupled with the international network of ideas is the solution to adapt to a more sustainable way of living. Maybe, thanks to open source development we can make a difference, instead of waiting that this solutions are developed in the closed of traditional businesses, and the important think is that we move in this direction, that is worth investigating.

Wikispeed takes the role to ensure this kind of products pass the regulatory testing that makes this products actually usable: without this our car wouldn’t even be allowed to go on the streets, that’s why we carry on safety and regulatory testing." (http://openalia.wordpress.com/2012/05/11/interview-with-joe-justice-of-wikispeed/)



Excerpt of the full interview conducted by Simone Cicero:

"Simone Cicero: First of all, we would like to know from you directly what’s the status of Wikispeed, and get some info about the collaboration with Open Source Ecology. Also, in this framework, how’s the Extreme Manufacturing platform and process growing these days.

Joe Justice: These are three questions back to back, let’s see if I can answer in an clear way. First of all, let me introduce Wikispeed in a one minute and a half version.

Wikispeed builds ultra efficient cars and we do this with seven days development cycles using agile methodologies. Those methodologies include several aspects, for example about managing distributed teams – like with SCRUM – or methods to ensure an high quality bar and focused work – much like Extreme programming and Test Driven Development as part of XP – that we reworked for the manufacturing process labeling it Extreme Manufacturing.

Traditional manufacturing runs in 3 to 25 years long development cycles: this means you can go to a Porsche dealer and buy a brand new Porsche 911 car and that would be the best engineers thought you might want 24 years ago and, if we stuck with this example, Porsche recently announced that the current Porsche 911 wil be with us for the next 14 years.

In Wikispeed we are aiming for mass customization, very rapid development and technologies and efficiencies that haven’t yet existed, that are fully gamechanging and not just and incremental evolution of old and sometimes defunct technologies.

To do that we iterate on seven days development cycles, that means that we can change every aspect of the car every seven days. This is possible through modularity: the car splits into eight modules that are loosely coupled so we can change one and not change the others.

Wikispeed is missionized to rapidly solve problems for social good. We don’t just make cars. I recently gave a talk on methods for vaccine distribution to help eradicate Polio: we worked with a group that develops low cost medical centers and communites for this and we’ve done significant work with them.

Currently, the status is that Wikispeed developed some efficiency proof of concepts: the best is a car that is secure and runs for more that a hundred miles per gallon but we launched an Indiegogo crowdfunding campaing (Ed: last night!) for Wikispeed to build commuter cars that people wants to drive every day and are ultra efficient and that can be manufactured at ultra low cost.

We do this with OSE (Open Source Ecology). OSE gives us the platform to share ours designs, our budgets, our build practices and our maintenance videos with the world so that people can build their own ultra low cost commuter cars.

Imagine in a world when your commuter car goes more than 100 Miler per Gallon in the United States cycles, that is about 1.5 liters per 100 km in Japan cycles for example, and imagine that you could maintain it by yourself – if you are inclined to do it – or that is very inexpensive to maintain if you’ve someone else to maintain it for you. Imagine it to be modular and changeable to follow your life changes, from a convertible to a sedan to a pick up truck, and that’s exactly what we’re bringing to everyone right now and OSE is giving us the platform to share it with as many people as possible.

XM is the metodology to allow other business to make this changes as quickly as wikispeed is prototyping. XM is an agile methodology, it takes the best methodologies as applied by the best software teams and extrapolates them to make them applicable to every industry. In particular we’re applying this to R&D, physical engineering and physical manufacturing and we think that such a process can be used for finance, insurance, energy, law, community management, residential and commercial construction and other businesses as well.

Extreme manufacturing takes the best practices for distributed team management, frugal engineering and frugal design and applies back to the physical world.

The status of that is available on the wiki of OSE as it evolves and explained in videos and I think we’ll have a book and set of lectures soon.


SC: ... “Who’s gonna finance this”. ... I mean, this is a community driven approach and your project is actually fueled by volunteers: understanding the role of “capital”, of financing, and the investment phase needed is key to understanding what you can actually produce with this approach.

[Joe Justice]: When you produce you need roughly three kind of things: materials, machinery, and person hours. We have volunteers for person hours, and we have donated machines and donated materials – or small amounts of capitals that led us to buy those. With the indiegogo capital campaign we are going to obtain more materials to be able to prototype more rapidly and increase our innovation pace, we are going to acquire faster machinery and that increases our pace of innovation again. We’re also going to be able to finance some graduate schools stipends – this is not a job, it pays too little for being it – that will allow this graduate school students to work almost full time with the team passionately and very quickly and this will end up in an increasing the number of iterations we can achieve, eventually increasing our pace of innovation once again.

So you see capital helps us to improve in all the three things that team wikispeed uses to move more quickly and if you think that we’re already moving at a pace of innovation that – in some ways measuring it – is 10000 % faster than industry norms you can understand what we’re talking about. And we have lots of fun doing it!


[Simone Cicero]: Amazing, we all believe that this wide efficiency differences will make this approach the winning one in the long term. The point is, thinking about the world in say five years, or at least in the future, when this will be potentially the de-facto standard, there will be way less profit in the world, and less traditional money involved in the production of goods isn’t it?

[Joe Justice]: The direction we’re trending is that the same amount of profit will be generated in the world but will be generated by everyone, in a way that is more equitable – in an equitable distribution of wealth mode.

So instead of us going into a traditional employer that keeps a larger percent of our work, and replicating the same approach with people that work under us – as a country we do it for counties that support us and they do the same for countries under them – instead of that, everyone will have an equitable share that is proportional to the amount of value he contributes to the world. We’re not there yet but we’re moving closer: open source hardware movements are making this more possible than ever before.

In addition, with these open source movements, we’re innovating more quicker and that’s compatible with the current financial structure: this allows all companies to have product development cycles that are much more quicker than they are now. But then what we are seeing independently from that, is the democratization of precision tooling: when people can buy machines – or build their own – that have a level of quality that is comparable to mass manufacturing. Suddenly you have the opportunity for a garage manufacturing revolution where people can make a new type of shoes or a new type of car and actually make it via viable competitive components in the marketplace today and that decentralization of manufacturing starts with decentralization of the supply chain and of profiteering.

And if the same level of profit exist in the world we have now the opportunity of equitably distribute that across the world, and the differences between a third world and a developed world country will then reduce.

Certainly there’s the possibility that traditional big business can act to impede that but we didn’t see it until now, on the contrary we’re seeing encouragement by these businesses and political bodies: we’ve seen poor protectionism movements and instead more embracement because rapid development products are appealing.


[Simone Cicero]: I’m not completely convinced that we can seek for embracement from the traditional economic structures since it seems to me that the inefficiencies in the production systems today have been crafted with the aim of actually generating profits from inefficiencies. The exploitation of inefficient work and your process are on very different sides: by increasing efficiency you make actually less work needed for doing things. Another implication of the theories and practices behing XM could be that there will be less paid work and much more socially contributed work and people will be much more involved in actually building and tinkering in the production of their own goods – so they will actually contribute “unpaid” work in directly into building their own things. This means that places where this kind of activities will happen locally are to appear in the future so probably there will be less factories and more labs where I can go and produce or repair things on my own. So the question is: how will these places be made, will they come out from the Fablabs/hackerspaces/makerspaces concept?

[Joe Justice]: When we have things like Arduino, that is absolutely Italian – so you see Italy, as well as every country, is leading this revolution – and we have hackerspaces where people can create with Arduino and sister platforms – such as netduino – in the field of electronics; and we have Fablabs where people can go and produce physical equipments we are coming closer to the capabilities of traditional manufacturing.

But there is a point you raised that I don’t see, at least in the nearest future: will traditional manufacturing disappear and will people be involved themselves directly in the manufacturing process of their own goods? There’s a trend towards localization and community availability of manufacturing that we haven’t seen since the trade guilds days – particularly in Europe – and with these available pieces of manufacturing like the Global Village Construction Set (GVCS), the hackerspaces and fablabs maybe someone in your community will build your next bread owen or window garden, your next cat feeder or your next pair of pants. It might not be you directly – since I’m not that savvy with sheet types, shape templates and so on to make excellent pants – but maybe someone in my community could go to the hackerspace or fablab and make a wonderful pair of cost competive pants and I might acquire those locally.

So I dont’ see me making all the things I need, surely some – I make my own car now! – and by making this open source and modular, someone else can make it without going through the landscape of all, for example, the automotive problems: they maybe can just make a better seat belt – by iterating only the interior module – and make it available to the public domain.

In other words you maybe will not be involved in producing your entire new car – at least not very soon – but everyone might be customizing it and some people maybe will be making some meaningful improvements and maybe everyone will be buying a locally crafted car that is made from locally sustainable materials in the local fablab. I do see this as a fairly near opportunity that will co-exist with traditional manufacturing and eventually be growing in market share.


[Simone Cicero]: Do you think that we will be able to build everything with this approach? I mean, when I think to consumer electronics – for example cellphones – or to the next quad-core chipset I see very complicated processes.

[Joe Justice]: Well, the GVCS, is made of fifty items, our car is just one of those, that represent all the items needed to grant the current level of comfort and livelihood, and once this will be complete – and this is being prototyped right now by OSE, headquartered in the Factor’e'Farm in Missouri in the US – people will be able to build almost everything. There’s not a machine to make cellphones in the GVCS but there’s a machine that can make CNC printed circuitboards and there’s a machine that can make injection molded cases and from those people can make their own cellphones.

I was reading a blog before you and I talk, about people using a CNC printer – that cost about three hundred dollars to build – to print chips, whit two different printing heads one for plastics and one for circuitry. Obviously we can still not print core i7 miniaturized size circuitry but we are getting closer to this and by thanks to the open source community we’ll be able to iterate and improve so I do see us, able to produce our next quad core CPU and I do see us able to reduce and reduce and reduce the amount of energy needed to do this.


[Simone Cicero]: Is there something we are used to have these days that is really hard to imagine to be operated in this cooperative and sustainable way. Let’s think about large chemical factories or planes, giant ships etc.. Is there any possibility that we’ll end up understanding that these kind of goods are not sustainable anymore and we should learn how to live without having access anymore to them in the future?

[Joe Justice]: From my experience and understanding I share your view. Many consumption trends and lifestiles we have developed will be subject to radical change in case we want to inhabit the planet for more years in the future. We have been around since 250 thousand years and when the first home sapiens came out there were less than one million of us and about the same number of chimpanzees. Now we have around 600 thousand chimpanzees and more than six billion humans. Maybe the number of sustainable humans on the planet is about one million or even less – given the changes in the environment that we made so far – and we have now an elaborate network of machinery and chemicals extraction to provide for that number of humans. Unfortunately those processes are based on resources that are limited or, at least, that we consume with a pace that is faster than the pace those resources are replenished. We need absolutely to develop more energy and resource efficient ways of production and maybe local production and local manufacturing coupled with the international network of ideas is the solution to adapt to a more sustainable way of living. Maybe, thanks to open source development we can make a difference, instead of waiting that this solutions are developed in the closed of traditional businesses, and the important think is that we move in this direction, that is worth investigating.

But back into your point, I don’t see us continuing living exactly as we do now, and replicating the exact same level for example of chemical processing or goods procurement, but I see the open source, local manufacturing and local sourcing processes making us able to iterate quickly enough that we may land in something sustainable in time to take care of our planet.


[Simone Cicero]: It seems that, to enable these transition, we are going to create a substantially huge body of knowledge: we must improve almost everything and my question is about whose the owner of this body of knowledge that we’re going to create collaboratively. How will we manage this library of commons that will be used across the world to produce goods that we’ll be using everyday? Do you think that we’ll see, as it happens for the software, for example with the Mozilla, Apache or Linux foundation, not for profit institutions taking care of these design process phase enable us to design the next wikiboat, wikiwashingmachine and so on? What’s the role of these players?

[Joe Justice]: If the information required to improve these materials is not publicly available and noone is allowed to access this information we simply wont achieve the advantage so, with regards to Wikispeed, in terms of availability these designs are totally available to anyone. On the other hand Wikispeed takes the role to ensure this kind of products pass the regulatory testing that makes this products actually usable: without this our car wouldn’t even be allowed to go on the streets, that’s why we carry on safety and regulatory testing. That could be a godd example of the role that these organization can have (Ed: Wikispeed is a for profit business although it follows a substantial ethic and revolutionary manifesto you can check it out here)

Apart from the design, the most important body of knowledge we’re building right now is the process itself. Imagine a world where everything we do takes 1% less time: waiting in the doctors office, filing our taxes, filling up gasoline in our car or in general, innovate in our lives.

XM, by mutuating processes by the most performing software teams in the world, promises even a little bit more and developing this process might be our greatest social good." (http://meedabyte.com/2012/05/09/interviewing-joe-justice-from-team-wikispeed-on-the-future-of-manufacturing-and-consumption/)

Discussion

What the software Industry got from the car industry is returned with interest

Charlie Rudd:

"It is well known that Scrum and XP practices, the cornerstone of Agile and Lean software development, were inspired by lean manufacturing methods, most notably as applied by Toyota. Since their introduction approximately fifteen years ago (Kent Beck’s Extreme Programming 1999, Scrum 1995, Agile Manifesto 2001), Agile methods have resulted in dramatic productivity gains throughout the software industry.

And along the way something else happened. As Joe explains it, since the design and production cycle-times for software development are typically much shorter than they are for manufacturing, respective incremental process improvement has often progressed at a more rapid pace. Consequently, there are instances where Agile and Lean methods in the software industry have surpassed their counterparts in manufacturing, the industry from which the originated. In addition, just as the software industry gained insights that led to huge waste reduction and quality improvements by applying the manufacturing metaphor to software development, so now is Wikispeed achieving transformational results in the automobile industry, by applying leading edge software design and Agile methods to car design and manufacturing.

In short, the Lean methods that crossed over from automobile manufacturing and transformed software development are now (through Wikispeed’s efforts) crossing back over and transforming the car industry (again).

The rapid innovation displayed by Joe and Co is a great example of, as Stephen Johnson puts it, “Where good ideas come.” As we recently discussed, Stephen Johnson makes the case that through the ages, most important innovations were the product not of single-minded geniuses but the byproduct of human collaborations.

Agile values are founded on the notion that innovation happens through collaboration. Stephen Johnson states this idea succinctly: “Chance favors the connected mind.” Wikispeed demonstrates to a broader audience what we in the software industry how have known for some time: Agile methods accelerate innovation." (http://www.solutionsiq.com/the-agile-ceo/bid/51480/Agile-Innovation-or-How-to-Design-and-Build-a-100-MPG-Road-Car-in-3-Months)