CIRP – Life Cycle Seminar 2002
Parametric Eco-Efficiency Analysis: A DfE Support Tool Joost Duflou, Wim Dewulf, Farid Al-Bender, Paul Sas, Christoph Vermeiren Katholieke Universiteit Leuven - Mechanical Engineering Department Celestijnenlaan 300B, B-3001 Heverlee, Belgium Tel: + 32 16 32 28 45, E-mail:
[email protected]
Abstract In this paper a methodology is described that allows providing pro-active DfE support based on functional requirements available in early design stages. The method, called Eco-PaS, is based on an expert system that contains basic knowledge of a range of design alternatives for different sub-systems. These are made available to the designer as functional blocks. Based on the functional requirements specified by the designer, appropriate technical specifications are determined. These allow an instant comparison between different design alternatives for each functional block. The proposed method has been implemented and tested for a range of elementary functions in the field of machine design, showing promising results. A number of examples illustrate the developed methodology. Keywords: Life cycle assessment, ecodesign, parametric analysis, Eco-PaS, conceptual design Introduction Increased environmental awareness throughout all levels of present-day society has led to an extended responsibility of manufacturers with respect to their products' environmental performance. As a result, industry has, over the last decade, started to increasingly apply the principles of Life Cycle Engineering and Design for Environment (DfE): the total environmental impact of a product can only be optimally controlled when considered already during product development. 1
A range of DfE support tools has been developed over the years. Figure 1 gives an overview of the major types of tools, categorised on the basis of two criteria: type of feedback versus time of application. The figure reveals one of the major limits of conventional DfE support tools: a clear void exists for tools making use of the functional parameters available in the early concept development phase of the design process. Two strategies can be followed to cope with this problem. One strategy is to rework generic guidelines in order to make them sensitive to the problem situation. This approach has been followed in the RAVEL project, as described in [1]: every guideline is assigned a set of meta data representing its usefulness in a user situation. Another strategy consists of estimating the technical parameters, needed as input for most assessment tools, as a function of the functional -1-
CIRP – Life Cycle Seminar 2002
parameters available in the concept development phase. This approach forms the basis of the Eco-PaS system introduced in this paper.
Figure 1 : Categorisation of DfE tools
The Principle of Eco-PaS The Eco-PaS (Eco-efficiency Parametric Screening) is developed as a pragmatic and timeefficient method for screening the eco-efficiency of technical solutions from the early design phases onwards. The Eco-PaS is based on the following principles: 2
-
A machine is usually conceived as an innovative combination of standard solutions to elementary functions. For example, the elementary function "illumination" is fulfilled by a selection out of the standard "lighting system" solutions.
-
The system requires input in terms of functional descriptions and constraints (called Functional Parameters) instead of technical parameters.
-
The system returns output in terms of quantified eco-efficiency performance indicators, even in the early conceptual phase.
-
The system uses a convergent approach: all solutions are considered to be viable candidates unless proven inappropriate.
The Eco-PaS consists of five modules, which are depicted in figure 2 and described in the following paragraphs: 1. a preliminary selection module for pre-selecting feasible solution principles, 2. a metric selection module for selecting an appropriate eco-efficiency performance metric, -2-
CIRP – Life Cycle Seminar 2002
3. a dimensioning module for parametric estimation of technical parameters, 4. a parametric eco-efficiency performance assessment module, and 5. a reporting and presentation module. It is clear that a parallel exists between the sequence of Eco-PaS modules and the phases of an LCA as standardised in the ISO 14040 series.
Figure 2 : Principle of the Eco-PaS (Eco-efficiency Parametric Screening) system
The system has been worked out for a number of typical functional blocks as applicable in machine design practice. In the following paragraphs, the individual modules of the Eco-PaS will be described, using the example of a well known functional problem: the isolation of rotational motion (i.e. a bearing selection problem). Preliminary Selection Module : Selection of Feasible Solution Principles The starting point of the method is the exact and abstract definition of the function required, e.g. the isolation of rotational motion, as well as the definition of the required performance and the constraints, e.g. maximum force F [N], rotational speed n [rpm], required life span Lh [h], and diameter of the rotating axle [d]. These constraints and performance requirements are referred to as functional parameters. The first input ("function required") serves as the basis for the EcoPaS system to consult a database with known solution principles. The technical feasibility of those solution principles on the basis of the given set of functional parameters is then assessed by an expert system. The resulting selection of potential solution principles can be modified through user intervention. The system is indeed not aimed at replacing, but at supporting the designer: a wide range of potential solutions is offered in order to widen the viewpoint of the user from his habitual pattern of thought and experience. In the bearing selection case, solution principles "rolling contact" (roller bearings), "air contact" (air bearings), and "sliding contact" (plain bearings) have been selected as a basis for further investigation. 3
-3-
CIRP – Life Cycle Seminar 2002
Metric Selection Module In ideal situations, attention has been paid to eco-efficiency aspects when defining and formulating requirements, such that eco-efficiency constraints form an integral part of the functional description. In that case, the selection of an appropriate eco-efficiency metric has occurred throughout the inverse supply chain, or is even part of sector wide industry standards. An example is the proposal for an eco-efficiency metric and for product oriented environmental performance indicators for the rail vehicle industry as proposed in [2]. 4
In many cases, however, eco-efficiency constraints are rather vague, and a holistic LCA/LCC based metric is chosen. A thorough description with respect to eco-efficiency indicator selection is, however, outside the scope of this paper. Though we are well aware of limiting the scope of the examples from eco-efficiency assessment to merely environmental assessment, the remainder of the paper will use the EcoIndicator99 [3]. Dimensioning Module: Estimation of Technical Parameters This module of the Eco-PaS system provides models to estimate technical parameters based on the available functional parameters. Technical Parameters (TPs) are parameters that uniquely, though supplier independently, describe a chosen technical solution: a technically skilled person should be capable of producing / buying the envisaged component based on the set of TPs. In case of a roller bearing, for example, technical parameters include the outer diameter [do] and the width b [mm]. 5
The models provided by this Eco-PaS module can be created by either (or both) theoretic or empirical modelling. In case of the roller bearing, the outer diameter, do is theoreticly modelled based on technical standards as di + fd * di0,9, with fd a complex function of the force F, the speed n and the required minimum life span L, such that do = f ( di, F, L, n ). The width b can, following the technical standards, be modelled as b = 0,5 * (do - di ) * 1, 15 = g ( di, F, L, n ). Eco-Efficiency Assessment Module: Estimation of Eco-Efficiency Performance The third module of Eco-PaS performs the eco-efficiency analysis based on three sets of parameters : technical parameters, system parameters and eco-efficiency parameters. 6
-
the technical parameters are obtained in the previous module, and describe the functional block under study,
-
the system parameters (SPs) describe the impact of the superior system, of which the functional block forms a part, on the eco-efficiency of the functional block,
-
the eco efficiency parameters - (E2Ps) represent the impact of the selected eco-efficiency metric on the total eco-efficiency score. Typical E2Ps are the EcoIndicator99 score of 1kg stainless steel or the EPS score of 1kg ABS.
The models can again be created by either (or both) theoretic or empirical modelling. Both methods are shortly exemplified.
-4-
CIRP – Life Cycle Seminar 2002
An empirical way was followed in the case of modelling the bill of materials for standard 3phase motors. Based on composition tables (between 5kW and 100kW), the amount of copper could in a straightforward way be modelled as mcopper [kg] = Pnominal [kW]. The maximum error using this formula is 20%, which is estimated to be very acceptable for an early estimation tool. Multiplying both sides of this formula with the E2P for 1 kg copper provides (part of the) desired eco-efficiency score. Theoretic modelling was used to estimate the energy consumption as part of the ball bearing's parametric model. Based on elementary mechanics and on the manufacturer's information, the environmental impact EI [EI99 points] caused by the friction loss can be estimated as: EI[EI99Points]=ì *F normal*(2ð*n[rpm]/60)*(d[mm]/2000)*(L[h]/3600)*EIen.type[EI99Points/J]/çdriv.syst.[%]
with
7
-
ì = 0,0015 (equivalent friction coëfficiënt),
-
EIen. type is dependent on the energy source of the driving system, and as such a property of the "superior system",
-
L = min {theoretic maximum life span of the bearing, life span of the superior system}. The theoretic life span of the bearing is a complex function of bearing geometry and applied forces,
-
çdriv. syst. is again dependent on the driving system. From a theoretical point of view, long discussions can be held about how to allocate the environmental impact of losses in the driving system between the driving system and (in this case) the bearing. For the remainder of the paper, this parameter is set to 100%, representing a full allocation to the driving system. Reporting and Presentation Module mPt
Eco-PaS EcoEfficiency
Technical Parameters
Functional Parameters
Figure 3 : Output of the Eco-PaS Prototype Software
-5-
CIRP – Life Cycle Seminar 2002
The results of the Eco-PaS can be presented in a number of ways. A direct result of using the system in a design situation is, of course, a comparison of the estimated environmental impact of different feasible design solutions for specified functional constraints. A prototype software implementation of the model provides the results as shown in figure 3. Towards future implementations, statistical information about the solution range as well as sensitivity analysis should be implemented to increase acceptability. A graphical representation allows to show the regions of the solution area where the different solution principles are appropriate (i.e. for example more than 20% better than the other solutions; see figure 4). Since a graphical representation is, however, limited to two parameters, the remaining parameters need to be held constant.
Figure 4 : Example of a graphical representation of the solution area for the bearing selection problem
Discussion and Conclusions The Eco-PaS tool described and illustrated in this paper allows to estimate the environmental impact of technical solutions based on functional parameters available in an early stage of the design process. Design concepts, described as functional blocks that are subject to functional requirements, are translated into technical parameters. These technical parameters are, in turn, used to assess the eco-efficiency of the solution. 8
Main advantage of the system is the sufficiency of an input in terms of functional requirements rather than the technical parameters needed by typical LCA applications. However, to increase the acceptability and usefulness of the system, the application of statistics and sensitivity analysis is needed. References [1] [2] [3]
Hermans K., et al, "Pro-Active Knowledge Support", in: Dewulf W., et al (ed.), Integrating Eco-Efficiency in Rail Vehicle Design, Leuven University Press (Leuven, 2001), pp. 121-131. Dewulf W., et al (ed.), "Using EPIs in Eco-Efficiency Calculations", in: Dewulf W., et al (ed.), Integrating EcoEfficiency in Rail Vehicle Design, Leuven University Press (Leuven, 2001), pp. 97-109. Goedkoop M., et al, The Eco-indicator 99 - A damage oriented method for Life Cycle Impact Assessment, Pré (Amersfoort, 2000). -6-
