P. 1


|Views: 1|Likes:
Published by Moulay Barmak

More info:

Published by: Moulay Barmak on Sep 08, 2013
Copyright:Attribution Non-commercial


Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less





Understanding Airlines’ Value Perceptions for Value-Based Requirements Engineering Of Commercial Aircraft

Xinwei Zhang a, Guillaume Auriol a, Hakki Eres b, Claude Baron a, Mario Kossmann c a INSA Toulouse, Université de Toulouse (xwzhang@insa-toulouse.fr, guillaume.auriol@insa-toulouse.fr, claude.baron@insa-toulouse.fr) b Faculty of Engineering and the Environment, University of Southampton (Hakki.Eres@soton.ac.uk) c AIRBUS (mario.kossmann@airbus.com)
Copyright © 2012 by Xinwei Zhang. Published and used by INCOSE with permission.

Abstract. Although Systems Engineering (SE) processes and standards are widely used in aircraft development programs, traditional requirements engineering practice for commercial aircraft does not explicitly address value perceptions and related, relevant information. In this paper, a novel approach is proposed to promote a better understanding of customer value perceptions for value-based requirements engineering. The approach is a four-step process leading from initial customer statements to component value models. A set of theories and methods is introduced in order to resolve the different concerns regarding the appropriate understanding of customer value perceptions, and based on these, the establishment of the requirements. A case study is used that shows the transformation of airlines’ initial expectation statements into a corresponding group value model, which supports the first two steps of the approach. The benefits of this approach are that perceived customer value can be explicitly qualified and quantified; and subsequently used reactively for design evaluation, and proactively for value-driven design.

1 Introduction
During the initial phrases of a new commercial aircraft program, many conceptual decisions are made, concerning primarily the early product positioning, the subsequent definition of top-level aircraft objectives (or technical requirements at aircraft level), and cascading these down within the extended enterprise. All these decisions are made on the basis of their assumed value contribution to airline customers and/or other relevant stakeholders. However, this assumed value contribution is more an implicit concept, rather than an explicit consideration in aircraft development programs. Value related information such as value dimensions and value drivers that allow for early conceptual orientation at the different development levels within the extended enterprise are not usually explicitly shared between the concerned organizations. Systems Engineering processes and standards are widely used in the aircraft development program, but they do not address “value” in much detail in the requirements engineering (RE) stage (EIA632 1998, De Chazelles et al. 2004). Airlines’ expectations and needs are usually collected and derived into technical requirements on the aircraft level in a value-implicit manner. Losing a holistic, valuefocused viewpoint is also obvious in the aircraft design and evaluation process (Jenkinson et al. 1999). For example, aircraft design alternatives are traditionally evaluated in terms of recurring and non-recurring cost by aircraft manufacturers, and

(2) Airline value models and their derived component level value models can be used to evaluate the available or potential commercial aircraft concepts. In Section 2. aircraft. It is a four-step process that is presented in light of the ‘multi-attribute utility theory’ and supporting methods. sub-systems and others) in the European aerospace industry and useful input during the development of the approach. and (3) airline value models and their derived component level value models can be communicated among different development levels within the extended enterprise. a novel approach is proposed that helps to better understand customer value perceptions for value-based RE. 2010). and simulated (e. Section 4 provides insights on the application of the first two steps of the approach to an example. The approach is applied in order to understand airlines’ value perceptions for the development of value-based requirements specifications of commercial aircraft. Evidently. the context of the approach development is introduced. One innovative aspect of the test case is to build a value model based on airliners needs. using the Monte-Carlo simulation) in the design and development of commercial aircraft. A schematic illustration of value-driven design among different levels of aerospace product is shown in Figure 1 (Cheung et al. and thereby enhance the decisionmaking process of selecting aircraft. 2 Context This development of approach to support value-focused RE builds on a current research conducted within the European Commission’s seventh framework (FP7) research project CRESCENDO in the domain of European aerospace industry in collaboration with mainly European universities and manufacturers.e. Finally. it is difficult to make beneficial value trade-offs between different dimensions and this will fail to provide high-value aircraft to airline customers and other stakeholders. It is a combination of customer value model and interface model which models the functional relationships between customer needs and top-level aircraft requirements. reliability. expectations and other statements paralleling with traditional RE process. This paper is organized as follows. The test case of requirements establishment and value generation provides a beneficial link to main manufacturers (i. All these dimensions are potentially perceived as valuable by customers and other stakeholders. This is shown as the system value model in the left yellow part of Figure 1. and there are other dimensions. safety. If too much attention is paid to one single dimension or some dimensions. and underlying theories and methods. cost is only one special dimension of value. system value model is a multi-attribute utility function rather than the originally surplus value based value model deployed in initial investigation.direct and indirect operating costs by the airlines or other operators. cabin comfort and others.. engine.g. In Section 3. With the system value model it is straightforward to calculate the utility of special inputs of aircraft attributes that are outputs of aircraft product model. In this framework. quantified (how much). the approach to support value-based RE is proposed with its process. Section 5 summarizes the conclusions drawn and future research directions identified. transforming airlines’ initial expectation statements into a group value model. which ensures the developed airplane reflecting customer value. In this paper. It is believed that the approach can contribute to the following three aspects: (1) Value can be explicitly qualified (what it is and what are its internally structural relationships). modeled (in terms of value models). with an emphasis on highly influencing value dimensions and value drivers for designing alternatives with high value perceptions. such as performance. such as means-ends analysis and part-whole analysis. Then evaluation of different alternatives and design .

such as subsub-component models. • How to construct component value models and sub-component value models. • How to construct aircraft value model. A schematic illustration of VDD in context of CRESCENDO Adapted from (Cheung et al. deletion of the part of “subcomponent model” will create an evaluation and optimization cycle from component models back to component models. the approach is proposed in . Pay more attention on intangible value dimensions rather than only on cost aspect (or surplus value). which includes: • How to construct airline value model. 2004. Ulrich and Eppinger 2007). Besides the above concerns. 2010) The problems in the processes are how value model can be built. This cycle is actually very flexible and models can be added to extend sub-component models to further finer models. Models can also been deleted from the cycle from subcomponent models to component models. 2008.optimizations is enabled in terms of customer value. This value-driven design (VDD) cycle in the Figure 1 is from subcomponent models through component models. Pahl and Beitz 2007. 3 The approach There are already a set of processes and methods available in literatures and industrial practices for RE (Agouridas et al. Figure 1. This capability is attributed to single attribute utility information and value trade-offs information contained in the value model. Foresee prototypes in place that can be integrated into industrial environments. De Chazelles et al. and system value model back to sub-component models. Bayus 2007. For example. As our intention is to focus on value at RE stage and not to depart from available good practice. E2. E3. and E4. discussions with industrial partners also help identify their expectations about approaching the development and test case application: E1. Establish value-driven traceability from customer statements to requirements. Enable value-driven trade-off capability at requirements level and design solution level. product model.

Fundamental objectives should be transformed into ECs that are of technical nature and influence the attainment of fundamental objectives. The underlying methods of this step are the relationship matrix of the quality function deployment and the response surface methodology. The underlying theory to support the quantification is the ‘multi-attribute utility theory’. A customer value model is then constructed based on the verified independence assumptions among the attributes. In this step component value models are derived from system value model through performing sensitivity analysis. In the next section. Figure 2. and a fundamental objectives hierarchy. an application of the approach is deployed for transforming airlines’ initial expectation statements into an airlines’ group value model. Attributes are specified to measure the attainment of fundamental objectives or customer needs. The essence of the approach is that our thinking in any product development decision should focus first on customer value perceptions and then on design alternatives that influence it. which focuses on the first two steps of the approach to illustrate its usefulness and applicability.The four steps of the approach 4 The application Step 1: Identify and structure objectives . The main mission of this step is to identify what is of value to customers in the current context and to create corresponding. The underlying methods to support this step are means-ends analysis and part-whole analysis. This includes the following four steps. which will later facilitate the process of quantification. The case study of steps three and four is under development and will be reported on in the near future. • Derive component value models. • Specify attributes and construct value model.parallel to the traditional RE processes and methods. This step is necessary to transform customer value model into system value model. • Transform fundamental objectives into engineering characteristics (ECs). optimization and evaluation. which are also shown in Figure 2: • Identify and structure objectives. logical structures of a means-ends objectives network. These component value models are used for component design.

and the answer may be that it directly influences profit levels. On the other hand. These transformed objectives may be fundamental. This simply highlights the importance of identifying fundamental objectives. maintainability and usability. . maintainability and usability to determine the degree of commonality. “Minimize fuel consumption” is a means objective. constraints. they are sufficient for the illustration of the approach. maintainability. such as operating cost and usability to route operation. which are not under the control of the current decision context of aircraft development. Table1: Part of initial customer statements. operating cost. it is found that it is difficult to make sensible decisions when commonality is considered directly. as they think it is one of the fundamental reasons to select a commercial aircraft.The initial airlines’ statements in the Table 1 are used as the starting point for deriving value model. A reckless combination of commonality. ECs. means or strategic objectives (Keeney 1992). and it may influence negatively “maximize performance”. objectives. such as the range and the maximum take-off weight of the aircraft. The objective of “minimize operating cost” is a fundamental objective for chosen airlines. such as “rationalize fleet”. which are fundamental objectives. which influence positively “minimize operating cost” and “maximize maintainability”. marketing and strategy. Then value trade-offs should be made among operating cost. The underlying reasons are that it influences positively the achievement of “minimize operating cost” and “minimize environment impact”. “Provide commonality” is also a means objective. performance and usability without distinguishing the underlying means-ends relationships will result in double counting the importance of commonality. It is possible to ask why “minimize operating cost is important”. The overall performance as a means influences several objectives. It is straightforward to transform the initial statements into objectives as shown in the third column of Table 1. Through means-ends analysis on “provide commonality”. An obvious fact regarding to the statements is that airlines try to express statements in different levels and granularities. The decisions of how much commonality to realize should all depend on its influence on operating cost. their transformed objectives Transformed objective may be strategic objectives. Although they are incomplete. such as pricing. goals. although there is a high frequency of mentions by airlines. profit is also influenced by other means. The airline statements may be in the form of needs. or design parameters. “Maximize maintainability” is uncovered as a missed fundamental objective in Table 1. The types and examples of initial airline statements are partly given in the first two-column of Table 1.

The first attempt of asking questions leads to a decomposition into .They are at a too high level to be controlled by decisions under aircraft development. In this figure. “How” questions should be pursued to explore its relationship with aircraft development. After a complete means-ends analysis on the objectives in Table 1. Figure 3. Table 2: Transformed objective and their fundamental objectives It is then necessary to perform part-whole analysis on the fundamental objectives in Figure 3. The fundamental objectives established from transformed objectives A fundamental objectives hierarchy for the objectives of “Minimize environmental impact” is given in Figure 4. Those uncovered fundamental objectives from the transformed objectives are presented in Table 2. the imposed constraints are usually to be utilized to check the technical feasibility and economic viability of the design alternatives. In this hierarchy. This will help to find the underlying fundamental objectives. a means-ends objectives network is established and a set of fundamental objectives is identified. a narrow perspective is adopted and environmental impact is limited to noise and emission impact that will be sufficient to illustrate the process and methods. which helps to clarify the understanding of fundamental objectives of high levels and granularities. which is a sub-hierarchy in the Figure 3. such as “Minimize operating cost”. The outcome may be that current fleet is in a certain situation in which longer range and more fuel-efficient aircrafts are needed. these transformations do not eliminate those original customer statements that are also useful for design. the objective of minimizing environmental impact is decomposed into a degree at which appropriate attributes for measurement can be selected. The same procedures are performed in other fundamental objectives. For example. However.

confidence is established that the set of fundamental objectives satisfies the desired properties of fundamental objectives specified by Keeney (1992). which is necessary to assign meaningful weights to fundamental objectives. Figure 5. . Asking further what questions lead to a fundamental objectives hierarchy of operating cost in Figure 5. Finally. Figure 4. Thus the procedure facilitates further quantification and value measurements. hard work and creative thinking on these objectives.direct operating cost and indirect operating cost. In order to enable transformation from qualification to quantification. A hierarchy of fundamental objective of “minimizing environmental impact” Through means-ends and part-whole analysis. A hierarchy of fundamental objective of “minimize operating cost” Step 2: Specify attributes and construct value model A hierarchy of airlines’ fundamental objectives provides well-structured forms of input for the quantification process. certain attributes or metrics are needed to measure the achievement of each of the fundamental objectives in the leaves of the hierarchy. a complete hierarchy of all the fundamental objectives is established.

noise depreciation index and monetary impact (Jenkinson et al. It is also possible to assess this value model directly from proxy attribute 𝑋1 to utility as shown in Figure 6(b). Mahashabde et al. “Mean maintenance man-hours per flight hour” is selected as an attribute for “Maximize maintainability”. Figure 6. to assess value model of attribute 𝑋1 in terms of “health impact 𝑋2). the overall health cost resulting from noise impact is hard to collect. for some objectives. A constructed attribute with four levels is used for safety measurement in Joint Airworthiness Requirements: Minor. Proxy attributes are then used as indirect measurements for fundamental objectives. While the difference. causes difficulty of verifying independence assumptions and assessing value models at later stages. These together give single attribute utility function over 𝑋1. one attribute is identified for each function in appropriate levels and for airplane’s capability to commit certain route. which is normally error-prone. it is difficult to assess the utilities of different noise levels. 1999. For usability of airplane. Then single attribute utility function over 𝑋2 is assessed. Hazardous and Catastrophic. it may be too difficult to identify direct measurements or collect the information for the direct measurements. For example. the difference of their utility cannot be obtained directly. say between 108 EPNdB and 90 EPNdB. which may be the direct measurement of some means objectives. e. but it demands more cognitive complexity to assess reasonable utilities of levels of proxy attribute 𝑋1.” meaningfully. Without clear consciousness of the relationships between noise levels and health impact. and these attributes are expressed as elements of a vector. However. attribute 𝑋1 effective perceived noise in decibels (EPNdB) is a proxy attribute for noise impact that maybe more appropriately measured by health impact (𝑋2). Possibly. Proxy attribute and its assessment of value .Seat mile cost measured in 2011 dollars is a natural attribute regarding to operational cost. However. For example. is obvious.g. “Dispatch reliability rate measured in %” is an attribute for “Maximize reliability”. a two-level process is implied as shown in Figure 6(a). EPNdB and NOx emission. it is modelled with a conditional probability function 𝑝(𝑥2𝑥1). Major. For example. Every level is given clear clarification in value so that customers or designers can easily find the safety level of a special aircraft. introducing of proxy attributes. The function between 𝑥1 and 𝑥2 is necessary to be found firstly. 2011).

Keeney shows that it is appropriate to determine an additive function form when the set of objectives are fundamental objectives and the objectives satisfy the set of desired properties of fundamental objectives (Keeney 1992.…. When additive independence is verified. direct and unambiguous. After specifying attributes. an additive function form can be used for compose together 𝑋1  and  𝑋2. For example. Necessary information for assessing one single attribute utility function includes: 1) range information of the attribute. the underlying process to determine it reasonably is following: (1) Collect range information of attribute. for all values of 𝑋1   and   𝑋2. Therefore. One hypothetical single attribute utility function over seat-mile cost is modeled and illustrated in Figure 8. it is a reasonable form for representing airline preferences. the customer is indifferent between the two lotteries.g. OM} in the leaves of the fundamental objectives hierarchy of airlines with a set of M attributes {X1. although it is time-consuming when 𝑀 is large. we assume that there is a set of M objectives {O1. And.g.   Figure 7. For 𝑀 attributes. operational. one attribute for one objective. …. we have confidence that additive function form 𝑢𝑥1. Single attribute utility functions for attributes are then assessed. Actually. although they introduce the difficulty of modeling or approximating the relationships between direct customer attributes and proxy attributes. XM}. We also assume that the selected attributes are all direct attributes and there is a transformation process from these direct attributes to ECs that are proxy attributes for fundamental objectives. and 4) certainty equivalence (or lottery equivalence). After carefully performing means-ends and part-whole analysis. e. multiplicative or additive function forms. which help finding function forms of the value model. e. to verify the relationships between seat-mile cost (𝑋1) and man-hours per flight hour (𝑋2). that is from 0. …. 3) risk attitude of the customer towards the uncertainty attainment of the attribute. it is time to verify possible preference independence assumptions among attributes. For single attribute utility function over seat-mile cost.03 to 0.. a test as in Figure 7 is deployed.𝑥𝑀=𝑖=1𝑀𝑘𝑖𝑢𝑖(𝑥𝑖) (1) is a reasonable approximation of customer preferences. there are oneto-one relationships rather than multiple-to-multiple relationships between objectives and attributes. These assumptions are reasonable. the same procedures are used.It is desirable that the set of selected attributes satisfies desired properties of attributes. which imply existence of additive independence between 𝑋1   and   𝑋2.1 dollar every seat-mile cost. that is. 2007). In order to facilitate latter illustration. (3) Verify airline risk attitudes . measurable. 2) the monotonicity of the utility function. Verification of independence assumptions Empirically. (2) Ascertain the direction of increasing preferences and monotonically decreasing preferences is verified.

1 dollar seat-mile cost. .towards uncertain attainment of seat-mile cost and risk averse is found as airline would prefer expected consequence to the lottery. Figure 9: Single attribute utility function: man-hours per flight hour These assessments of single attribute utility functions are implemented in the Vanguard Studio.03. which includes six possible sub-models: • Increasing risk averseness model. A Vanguard Studio model has been developed for the assessment. which is one of business software for decision analysis and has a wide range of applications in industries and by our partners. and 0.1> which yields either a 0. and (6) Solve a. which is a linear form. (5) Assess certainty equivalence (or) lottery equivalence. each with a one-half chance.075 dollar is determined as indifferent to the lottery <0. b and c in equation (2) for determining single attribute utility function over seat-mile cost using available information. Single attribute utility function: seat-mile cost The single attribute utility function over man-hours per flight hour is assessed of risk neutrality as shown in Figure 9.5. and the value of c indicates the degree of airline’s risk averse. constant c<0 indicates monotonically decreasing preferences.03 dollar or a 0. 0. (4) Select the class of risk averse utility functions (Keeney 1992): 𝑢𝑖𝑥𝑖=𝑎+𝑏(−𝑒−𝑐𝑥𝑖) (2) where a and b>0 are constants and are selected to ensure ui being scaled from zero to one. Figure 8. 0.

customer’s risk attitudes and assess the single attribute utility automatically.𝑥𝑀=𝑢(𝑥1. then 𝑢𝑥1.…. • Increasing risk proneness model. For example.0.0.04. Then the model can judge the direction of increasing preferences.5 (4) 𝑢𝑖𝑥𝑖′′′=𝑎+𝑏−𝑒−𝑐𝑥𝑖′′′=0 (5) These three equations have three parameters and can be easily solved in some mathematical software. such as Maple. which shows one equation with ki’s as unknown. For example.• Increasing risk neutrality model.1. But it poses problems to the Vanguard Studio. if 𝑥𝑖′′<𝑥𝑖′+𝑥𝑖′′′2 could be further found. three equations can be found.0. Actually.…. then the single attribute utility function is increasing. with two hypothetical alternatives of the same utilities for airlines differing only in achievement of two attributes: seat-mile cost (Xi) and manhours per flight hour (Xj).03.2 man-hours per flight hour. The inputs of the model are following (when monotonicity is verified): • The most desired level of the attribute 𝑥𝑖′. • The minimum acceptable level of the attribute 𝑥𝑖′′. If an airline would like to pay 0. • The mid-value point in the attribute range 𝑥𝑖′′′. if 𝑥𝑖′>𝑥𝑖′′′ could be found.𝑥𝑀) (9) is satisfied. .…. after some transformation. (7) and (8) that could be solved in Vanguard Studio.3.0. The underlying rational is that when there are the inputs of 𝑥𝑖′. which is visually displayed in Figure 10. Weight ratio between kj and ki is shown at the bottom of the figure. 2𝑒−𝑐𝑥𝑖′′=𝑒−𝑐𝑥𝑖′+𝑒−𝑐𝑥𝑖′′′ (6) 𝑏=1𝑒−𝑐𝑥𝑖′′′−𝑒−𝑐𝑥𝑖′ (7) 𝑎=𝑏𝑒−𝑐𝑥𝑖′′′ (8) Scale constants (or relative importance) ki’s are assessed through making value trade-offs. • Decreasing risk neutrality model. • Decreasing risk proneness model. these equations can be transformed into three single-variable equations (6). • Decreasing risk averseness model.…. then the customer is risk averse over the attribute.….  𝑥𝑖′′  and  𝑥𝑖′′′.…. which gives the following equations: 𝑢𝑖𝑥𝑖′=𝑎+𝑏−𝑒−𝑐𝑥𝑖′=1 (3) 𝑢𝑖𝑥𝑖′′=𝑎+𝑏−𝑒−𝑐𝑥𝑖′′=0.01 dollar more in seat-mile cost in order to exchange a reduction of 0. as there are no available functions for solving nonlinear equations.

we should consider the case of two airlines. In this case. a multiattribute value model for individual airline is established. single attribute utility functions and ki’s. a set of at least M equations is constructed to determine the M ki’s in the additive function form. a sophisticated assessment process is necessary. Some attributes may be assigned a higher weight by a certain airline while others may assign them a lower weight. They then may have different value models. Airlines may even have different perceptions about single attribute achievement. market potential. These equations are linear and can be easily solved in Vanguard Studio. However. Value trade-off to determine relative weights For M attributes. different airlines may have different preferences about attributes and their importance. political considerations and others. they have different preferences about relative weights and attainment of objectives. Table 4: A formulation of group utility problem The two airlines with three attributes can be easily extended to N customers with . the Supra Decision Maker (Keeney 1993) is the aircraft manufacturer who verifies the assumptions and assesses the scaling constants ws that address interpersonal comparison of utility. The roots of these equations are the ki’s. which is another decision-making problem of multiple objectives. In order for an aircraft manufacturer to meaningfully assign relative weights of airlines. Table 3: A group utility problem A formulation of the group utility problem in Table 3 is given in Table 4. After assessing function form for the multiple attributes. The weights are influenced by percentage of orders. If consensus cannot be achieved. In order to simplify the illustration. However. One airline is a low-cost carrier and the other is a flagship carrier. This example is demonstrated in Table 3. although the same set of fundamental objectives and attributes is used for assessment.Figure 10. They have the same set of fundamental objectives that are narrowed down to three in our context. group preferences have always to be derived.

  Figure 11. Parts of them are shown in Figures 12.. sensitivity analysis. These modeling and simulation have been also implemented in the software of Vanguard Studio. Group utility and color-based sensitivity analysis Other capabilities of value modeling and simulation are also implemented. Individual utilities are assessed and aggregated meaningfully into group utility without making compromise between airlines.M attributes 𝑢𝐺=𝑠=1𝑁𝑤𝑠𝑢𝑠𝑥..    𝑀 (10) With a quantified group value model and several individual value models. . it is then possible to do various kinds of visual analysis.. which shows the degree of relevance between customer attributes and group utility. different perception of attribute attainment and different relative weights of attributes. Differences between airlines can be identified: different relative weights of airlines. which supports distributional design and development. Monte Carlo simulations and optimization.  . This value model can be conveniently integrated into existing engineering models in an Isight simulation process workflow when the interface between top-level aircraft requirements and customer attributes are available. such as component-based modeling and internet-based simulation. Figure 11 represents the implemented model of group utility function and a color-based sensitivity analysis.  𝑖=1.

These theories and methods were introduced to resolve different concerns appearing during the different steps of the approach. which offers significant potential to enable the development of aircraft that are perceived by customer airlines to be of high value. This value-based information can be used for the conceptual orientation at all different development levels. for major modification projects or new aircraft programs. quantified. and reduces the ‘time to market’ of new aircraft. when the necessary information of aircraft concept alternatives is collected. the configuration management of the related value-based information at all levels concerned. With the approach and the illustrated case study. The future research should focus on a number of aspects regarding the early and ongoing interaction between value models and simulations across multiple levels of development within the extended enterprise. • Value dimensions and value models can be explicitly shared at the different development levels within the extended enterprise. Value modeling and simulation 5 Conclusions In this paper we proposed an approach to understand customer value perceptions for value-based RE and a case study of transforming airlines’ initial expectation statements into an airlines’ group value model. value-based RE can be integrated into existing quantitative engineering models. This is a very attractive feature that enhances the decision-making process. ‘partwhole analysis’. A scalar value is calculated for the set of information attributes of a specific aircraft alternative. and the re-use of this information.Figure 12. including traceability information. The approach is a four-step process integrating the existing ‘multi-attribute utility theory’. ‘response surface methodology’ and others. the approach reduces development risks throughout the extended enterprise. Therefore. ‘means-ends analysis’. modeled and simulated. Therefore. • Airlines’ value models can be used to evaluate available or potential aircraft alternatives. when the validated technical requirements at these levels are not available yet. This information was used to enable value-based RE of commercial aircraft. which was presented through the case study in Section 4. . we identified the following benefits: • Airlines’ value perceptions can be explicitly qualified.

. Decisions with Multiple Objectives: Preferences and Value trade-offs. and von Winterfeldt. optimisation methods.. Ulrich. and value-driven design... J. S. Biography Xinwei ZHANG is a PhD at LAAS-CNRS laboratory. K.” In: Shane. 1998. Cambridge University Press. He has published several papers in leading international journals.” Paper presented at the 14th Annual International Symposium of INCOSE. P.. and Anne. Civil Jet Aircraft Design. EIA (Electronic Industries Alliance). Hakki ERES is a senior research fellow at the Faculty of Engineering and the Environment. 4th edition. Cambridge. 1992.” Requirement Engineering.. University of Southampton. degree in 2004 to develop a master degree program in system engineering... Forrester.. He joined the INSA Toulouse after his Ph. He obtained his PhD degree in Mechanical Engineering from University of Delaware.. B. M.References Agouridas.. Bayus. where he teaches system engineering. 2012. 2011. G. Product Design and Development. 2007... Hollingsworth. Keeney. Winand. Claude BARON is professor in computer sciences at the INSA Toulouse (France) in the department of Electrical Engineering and Computer Sciences. Processes for Engineering a System ANSI/EIA-632.. Value-Focused Thinking. computational fluid dynamics applications...R. J. Cambridge University Press.” AIAA Journal of Aircraft (in print) De Chazelles. Miles.. He received the M. J. 47(1):15-52. D. Her current research focuses on system engineering in the LAAS-CNRS laboratory. AIAA Press. D. S. 13(1):19-48. A. He received his BSc and MSc degrees in Mechanical Engineering from Bogazici University. Advances in decision analysis. A. Turkey in 1990 and 1993. S. 2007. H. France.. J. W. Engineering Design: A Systematic Approach. McGraw-Hill/Irwin Press.. W. 2008.. P. P. eds.. R. Mahashabde.. 3rd edition. “Understanding Customer Needs.” Progress in Aerospace Sciences. Guillaume AURIOL is an associate professor at INSA Toulouse. 1999. Cheung. in 2003. “Application of Value-Driven Design to Commercial Aero-Engine Systems. R. degree from ENSICA (Toulouse). L. Jenkinson.D. Pahl. MA. Scanlan. and Rhodes. H. K. H. “Advanced Product Planning: A Comprehensive Process for Systemic Definition of New Product Requirements. Comes. and does his research in the LAAS-CNRS laboratory. P. V. Wiseall. S. Jr. and Eppinger. 1993. and von Winterfeldt. R.. A. Harvard University Press.. L. Simpkin. INSA Toulouse.. His research interests are in the fields of numerical solutions of coating flows. L. and Briceno. . R. L.. Keeney. D. He is a member and an activity leader of AFIS (affiliation to INCOSE). Eres.. Wong. and Raiffa.... USA in 1998. 2007. et al. a French graduate school of Aeronautics. Springer. Cambridge.” In: Edwards. His research aera is requirements engineering for complex systems and he actively paticipates a European project CRESCENDO to develop approaches for value-based requirements engineering. “Practical Value Model. eds... EIA 632.. 2007. and Pennington. 2004. respectively. “Assessing the Environmental Impacts of Aircraft Noise and Emissions. Collopy. MA: Blackwell Publishes. Blackwell handbook of technology and innovation management.E. “Customer Focused Engineering in Airbus A380 Programme. Keeney. McKay. and Beitz.

He has served as a naval officer with the German and French navies. in Requirements Engineering from the University of the West of England (UK).eu) under grant agreement n◦234344. Technical Manager and also Consultant in Services Marketing. . and Ontology-driven Requirements Engineering (OntoREM). an MBA from the University of Warwick (UK). and was awarded an MEng in Aerospace Technology from the University of the Federal Armed Forces in Munich (Germany). Currently. Acknowledgements: We thank Anne Monceaux (EADS Innovation Works) for discussions about value modelling and simulation at requirements stage. The research leading to these results has received funding from the European Community’s Seventh Framework Programme (FP7/2007-2013) (www. Mario is the author of the book ‘Delivering Excellent Service Quality in Aviation’ (Ashgate 2006).D. having previously worked for Blohm & Voss as a Systems Engineer. and a Ph. Product Family Management (PFM).crescendo-fp7.Mario Kossmann is an experienced Systems Engineer and Capability Integrator for Airbus. Mario works as Systems Engineering specialist and Product Development Process Architect for Value-Driven Design (VDD).

You're Reading a Free Preview

/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->