Tuesday, October 29, 2013
ETRIA TRIZ Futures Conference, Paris. Day 1 afternoon.
We had a short walk to the Paris City Hall for a beautiful lunch -- 3 courses with 2 wines! The afternoon speakers will have a challenge rousing the audience. Good news, there was lots of afternoon excitement.
Val Souchkov was the chairman of the afternoon session on
TRIZ with other tools. First speaker
was Ms. Yui Kato with Manabu Sawaguchi, “Design Process Management based on
Redesigned Contradiction Matrix in Aesthetic field.” She gave a detailed explanation of the issues
of design workflow and individual designers’ differences that contribute to the
need for improvement in design work.
This study is limited to the product design work phase of product
development (excluding market research and product planning that come before,
and production and distribution that come after.) They have hybridized the stages of design
based on AHP (especially weighting and evaluation) and the stages based on TRIZ
(design and problem solving). Improvements
to designs that have a good sales history are different in some details from
new products. A typical esthetic
conflict was improving Impact vs. declining
Color Harmony; a sample of 11 elements of an aesthetic issues matrix was
shown, with a 48-element version
described that included some of the classical Altshuller matrix elements. Future work will include extension of the
examples of application of the principles to the design problems.
“Lean-TRIZ instead of TRIZ Lean” was introduced by Prof.
Christian Thurnes (the lean guy) and Dr. Frank Zeihsel (the TRIZ guy.) They start with the assertion that lean methods have had a lot of money and
management attention over the last 30 years, whereas TRIZ has been an expert
method with very little management
attention. They have enriched TRIZ tools with lean elements, and then used them
to work on lean problems. For example,
they have made specific separation principles that include separation of
non-value added work, separate set-up tasks from operating tasks, etc. Likewise, ideality was easily combined with
lean—what is the ideal production rate, what is the ideal transportation unit,
…what are the ideal functions of the product, ….Then they reversed the process
and found areas of weakness in Lean that can be improved with TRIZ, such as
treating “necessary waste” as a physical contradiction, then applying the usual
separation principles to resolving the contradiction. They
conclude that these practices have a high potential for spread in the lean
world, with case studies and stories needed to support the dissipation of the
methods.
I presented the paper written with Tim Brewer, Joe Miller,
and Darrell Mann on the crowdsourcing and crowdfunding business models, and how
the TRIZ model of the complete system helps to understand these emerging
models. I’ll post the full paper on
this site when I get back from the meetings.
Prof. Paul Filmore presented the work done with Mir
Abubakr Shahdad on the application of
TRIZ to graphic design, using genetic algorithms. They developed software, now in its 7th
generation to create a computer-aided innovation tool. Genetic algorithms are used because of their
ability to deal with extremely large search space. Font design was the specific graphic design
task that was chosen. The designer
selects “chromosomes” and the system generates the next generation; the designer selects the “parents” of the next generation, and the
system does the iterations. A similar
process is applied to logos. Feedback
from the designers is that they find it very fast, works well for initial
concept creation, and enables them to benefit from TRIZ without studying
it. New studies will include examining
the difference between the use by graphic designers and engineers, and
inclusion of additional TRIZ principles.
“A novel hybridized TRIZ-based Design Approach for Concept Generation”
was presented by Aiman Ziout and Ahmed Azab from the University of Windsor,
Canada. They demonstrated the method
with the design problem of an active joint, one that can be disassembled by some
external trigger. Active joints are
important at the end of product life, for disassembly, as an element of
sustainable design. Function analysis
and patterns of evolution were combined with Cladistics (a biological method of
classifying solutions) to create lines of ideas.
Valeri Souchkov presented “Trend of Functionality Evolution” which is
based on some of Altshuller’s work, expanded by his personal experience with
over 100 products and projects. During
his work, Val realized that something was missing in TRIZ, and he used
TRIZ-based function analysis and value analysis to complete the missing
element, the evolution of the functionalities.
He saw 18 steps of evolution, 11 of expansion and 7 of convolution. He called on his experience in the The
Netherlands to use the evolution of the bicycle to illustrate the 18
steps. Business-targeted products (B2B)
have a somewhat different emphasis from B2C products, and the order of the
patterns is not always the same. Val’s
challenge to the audience is to use these trends of functionality evolution to
support innovation road mapping and risk management.
Eric Prevost’s presentation “TRIZ for Business” emphasized the need for different vocabulary
and different psychology than those used for engineering/technical TRIZ.
His first approach is to create fear—show the statistics on the vanishing
of companies (87% of the Fortune 500 changed in 20 years…) and to show new paradigms
in technology and manufacturing, as well
as new paradigms of selling and other elements of business – sensors in tires
and engines and many products make it possible to sell based on performance and
lifetime rather than price. Innovation
models are changing—particularly integration of service into products
accelerates innovation, but requires new business models with new
partners. As an example, he showed a
conventional business process used at CapGemini, then inserted TRIZ into the
process, and did 3 one day workshops (sales, talent management, and project
funding.) Result: lots of ideas, reusable content, and great
TRIZ interest.
Val Souchkov presented Vladimir
Petrov’s paper “On Su-Field Analysis for Information Processing
System” which was originally presented at
the TRIZ Developers Summit. Petrov
replaces the Substance-Field interaction
with DFK (Data coming into the system,
Function that changes the data, and
Knowledge that is the aggregation of information about the incoming data
and the function, and can be used to adjust the system. ) There are 3 laws:
1.
Multistage
processing: 4 trends within this trend
progress from simple to simple multistage, to coordinated multistage (knowledge
about stages interacts, improving system function) to common multistage which can have shared structure with shared
knowledge.
2.
Multiple sources of processing: Independent processing of multiple kinds of
data progress to coherent multiple source processing, (video conference example, which video and
audio data are processed separately, but knowledge about positions of people is
used to modify the process and the data collection.)
3.
Accommodation:
And information system tends to accommodate past data to improve its
performance. The progress is from
static, to learning, to evolving DFK systems.
The function stays the same but the knowledge is used differently at
each level.
Complex systems can be generating by combining systems from each of
trends.
The meeting adjourned for the day, and we found congenial small
restaurants for continuing the discussions.
