Eco-innovation Manual: a step-by-step
methodology for eco-innovating products and
services.
Robust Analysis and concept development crank
mechanism.
Capturing design intentions when working in a
virtual environment.
Implementing eco-innovation into SMEs in the developing economies
In 2017, the United Nations Environment
Programme (UN Environment) and the Technical
University of Denmark released the Eco-Innovation
Manual as a result of a four-year initiative.
The Eco-Innovation Manual introduces a
methodology of eco-innovation within small and
medium sized enterprises (SMEs) in developing
and emerging economies. The primary goal is to
support manufacturing companies to improve
their sustainability performance within innovation
and product development by providing a
step-by-step approach that aids their development
activities.
The Eco-Innovation Manual includes a number
of complementary materials: (1) Business Case
for Eco-Innovation (with good practices to demonstrate
the case for eco-innovation); (2) Sector
Supplements (for the Chemicals, Agri-food and
Metals sectors); and (3) Templates (to support
the implementation of the proposed tools and
activities).
The manual was co-developed in collaboration
with a number of stakeholders (including
industry, academia, government, and nongovernmental
institutions) from 76 countries.
Currently, the Eco-Innovation Manual is being
implemented and deployed, and a number of
success case stories start to emerge worldwide.
Eco-innovation is the development and application
of a business model shaped by a new business
strategy that incorporates sustainability
throughout all business operations based on life
cycle thinking and in cooperation with partners
across the value chain. It entails a coordinated
set of modifications or novel solutions to
products (goods/services), processes, market
approach and organizational structure which
leads to a company’s enhanced sustainability
performance.
More information about the project, description
of case studies and download of the manual
and complementary material can be obtained at
http://www.ecoinnovation.org/
Rethinking Mechanical Designs - Robustness of combustion engines
Combustion engines have a well-known design
which has proven its functionality in innumerable
applications for well over a century. At the
same time, the design of a combustion engine is
largely relying on a number of overconstrained
parts and interfaces, which require extensive
quality control efforts in practice. From a theoretical
perspective, it is consequently largely
unpredictable/unrobust.
It is one of the central promises of applying
Robust Design that it allows us to re-think and
optimise corresponding mechanical designs
that are almost universally adopted and rarely
questioned. To investigate this potential, the
DTU Robust Design Research Group conducts
systematic Robust Design analyses of widely
implemented design solutions, for example of
automotive assemblies such as gearboxes and
combustion engines. Focusing on an integrated
consideration of different variation sources in
production, assembly, and use, the studies have
provided important insights into the applicability
of Robust Design in a practical context, and
yielded proof-of-concept solutions that offer an
enormous potential in terms of improved performance
and/or allowable tolerance windows.
In the future, these initial results will allow
the further development of Robust Design
principles and methods, as well as boost the
use of design optimisation tools for ensuring
the mechanical robustness of products, devices,
and production equipment.
What are the prospects for AI in Engineering Design?
Staff from the Engineering Design and Product
Development section presented their work at
CAE Inc. in Montreal, Canada, on the occasion of
the Design Society’s Advisory and Management
Board meetings. In a wide-ranging review of
developments in design informatics, they asked
what were the prospects for developing engineering
systems that could genuinely learn - for
example how to build a good analysis model
from CAD, what the best-practice costs are for
different manufacturing processes or what are
an engineer’s design intentions while working
in a virtual environment? The pre-requisites
are to have access to sufficient examples of
readable models for patterns and features to
be identified and then to be able to map in a
computer-interpretable way the dependency
relationships between the information objects
generated in the design process.
Today, the most stunning recent developments
in the applications of artificial intelligence (AI)
are taking place using unsupervised learning
using enormous data sets. The barrier to this
taking place in engineering is that most engineering
data is proprietary in ownership and often
also in data format, and that in engineering
we use a wide range of highly interconnected
models to represent different engineering viewpoints.
To be able to exploit such data, beyond
what can be learned in a single organisation, we
need mechanisms to allow data to be shared for
learning purposes while maintaining confidentiality,
and to redouble efforts to ensure that
engineering data are interoperable.
Contact:
Tim C. McAloone, e-mail: tmca@dtu.dk
Contact:
Tobias Eifler, e-mail: tobeif@mek.dtu.dk
Contact:
Christopher McMahon, e-mail: chmcm@mek.dtu.dk
28 Engineering Design and product development