An investigation into how structural design process performed by industrial design engineering trainees unfolds

Proceedings of the ASME 2012 International Design Engineering Technical Conferences & Proceedings of the ASME 2013 International Design Engineering Technical Conferences & Computers and Information in Engineering Conference IDETC/CIE 2013 August 4-7, 2013, Portland, Oregon, USA

DETC2013-12318 AN INVESTIGATION INTO HOW STRUCTURAL DESIGN PROCESS PERFORMED BY INDUSTRIAL DESIGN ENGINEERING TRAINEES UNFOLDS Eliab Z. Opiyo Faculty of Industrial Design Engineering Delft University of Technology Delft, Netherlands Email: [email protected]

ABSTRACT This paper presents a study conducted to investigate how structural design process performed by industrial design engineering trainees unfolds. The focus was specifically on structural design of industrial design engineering products, a class of consumer durables designed to endure short and long term regular usage by human beings. The study consisted of a questionnaire survey and interviews, as well as observation of structural design processes. One hundred and eighty two subjects participated in the questionnaire survey. The study gave us some clues on how structural design processes proceed in real life and on how a formal method for structural design may ideally be structured. Furthermore, the results of this study suggest that it is not always important to follow a strictly structured order in executing certain structural design activities. Specifically, sequential execution of the mid-way activities of materials selection, process selection and engineering analysis is somewhat impractical. A linear-cyclic structural design procedural process model is composed based on the findings of the investigation. However, since the study involved only a handful of selected structural design processes, we cannot draw out definitive conclusions regarding the applicability of the composed process model or suggest how strict the process steps must be adhered to. Further studies are needed to verify the validity of the findings of this investigation. INTRODUCTION In today’s product development processes, there is often the need for formal structured methods for guiding the designers and engineers in performing product development activities – see, e.g. [1]. It is widely acknowledged in some literature that the process models used in realization of products influence qualities of final actual products – see e.g., [2-3]. And

some researchers in the recent past decades have directed their efforts towards developing models of design process (also known as phase models or procedural models in some literature) to help guide activities within the product development interval. Many structured process models have been created and some of them have already been put to use. For instance, some approaches for systematic design such as those proposed in [4-5] have come a long way over the years; and are tacitly used in practice in one way or another. However, there have always been some fundamental issues that need attention, including, for instance lack of flexibility when it comes, e.g., to modifying semi-finished designs and the inherent strictness of the process models when it comes to promoting creativity. In general terms, there are uncertainties as to whether the existing process models for guiding the execution of activities are good enough, if they really work and help improve qualities of products as anticipated, and as to whether the granularity of the descriptions of the process models are detailed enough. This study partly focused on the granularity issue and we specifically investigated the structural design process1 with a view to uncover if there is a need for a systematic process model for guiding structural design activities, and if so, how should it be like and what should be the process steps. Overall, we argue that the existing process models are coarsely defined and do not scale down to address some of the challenges in structural design. We are not aware of the existence of any formal procedural process model created specifically for guiding structural design activities. The importance of having well-organized design process activities (including the structural design activities) and the significance of the decisions and choices made during the 1 (- part of the design process that follows after concept design and proceed detail design and prototyping)

1

Copyright © 2013 by ASME

(or testing) and use (or operation) as the main phases. During the specification and analysis phase, goals are identified and the requirements, which state what the product will do and be like, are specified. In the design phase, sub-components of the product or system are identified, the goals are decomposed into targets, and the final design, which depicts how the product will be like and how it will be manufactured (or implemented), is prepared. Afterward, the components are produced (or implemented in case of software products) according to the designs and the components and sub-assemblies (or units of software product) are tested as they are completed. During the inspection and testing phase, classical tests such as reliability tests, performance tests, benchmarks and stress tests are conducted to determine whether the previously specified requirements have been met. It is important to note that various kinds of models typically offer the designers and engineers with alternative ways of accomplishing the development activities. In short, the prevailing process models only present the common stages of the development process, without prescribing, for instance, the activities that must be performed within the phases or the order of execution of the activities. Although there is a general consensus about the names of the phases of the process models, there is however a notable disagreement about the actual activities and the deliverables of each phase. Moreover, process models typically have certain inherent properties. For instance, some of the models are linear, others are recursive, and some put much emphasis on the description of the early phase activities. The influence of formal methods (especially in the processes of development of software products) has been studied extensively by some researchers - see, e.g., [2], [14]. One of the limitations of the existing formal methods for guiding the product design activities is that they are coarsely grained and do not scale to address flows of activities and quality improvement challenges within the sub processes. Overall, most models are considered to be inadequate on their own largely because the granularity of the process steps in them is too large. The emphasis of life cycle descriptions is usually on outlining abstractly the guidelines for engineering the products, but these descriptions fail to specify the elementary building blocks necessary for managing and coordinating the product development activities. Therefore, we argue that there is need for a crispy formal method for guiding the activities within the sub processes and for improving the processes and quality of the in-process deliverables. We argue that the better the quality of in-process deliverables, the better the chance of acceptance of the final product. Therefore, since the general understanding is that formal methods contribute to the improvement of the development processes as well as of quality of the final product – see e.g., [3], [15], it follows naturally that the application of formal structured methods in structural design processes would also have similar effects, namely; would also improve structural design process and quality of the in-process deliverables. Presently, the truth on ground, however, the existing process models are excessively coarsely defined (– see [4-5], [7-9]) and

design process on quality and on the acceptance of the eventual product is well documented. It is widely known that quality of a product is partly determined at the design phase and that a large proportion of product costs is usually committed at the design phase as well. Structural design is the sub process within the design process in which materials are selected, preliminary manufacturability decisions and choices are made, engineering analyses (such as analysis of strength and stiffness of structural members) are carried out, and the geometry – including the dimensions of the components are defined in such a way that the prescribed requirements for the product and the desirable physical behaviors of the product are realized. The decisions and choices at this stage determine the material quantity and production processes - see e.g., [6] and therefore strongly influence quality and the eventual cost of product, and impact the environment as well. It is therefore imperative to have in place mechanisms for ensuring that the structural design process proceeds in an orderly way, all activities are performed, and that correct decisions and better choices are made. In this work, we investigated how structural design process in industrial design engineering unfolds. Based on the findings, we concluded on whether or not it is worthwhile to have in place a formal method for guiding structural design activities, and if so, how should it be like and how the activities should be organized. The paper is structured as follows. We present a concise literature review and describe the research problem and hypothesis in the subsequent section. We then describe the methodology used in the investigation, present research results, and discuss these results. A CONCISE LITERATURE REVIEW AND PROBLEM ANALYSIS It is widely argued in the literature that application of formal methods, both in the specification and in the design processes provide further quality improvements through reduced defects in the final products [2-3]. Many process models for guiding the processes of development of engineering products have been proposed in the past, including for instance, the phased process models proposed in [4-5], [7]. Phased models split the design process into multiple phases and typically suggest sequential flows and execution of activities between phases (although in practice this often is not necessarily the case - i.e., one phase is not necessarily bound to be finished first before the next one is started). Other process models such as those proposed in [8-9] demand the design process to navigate randomly between domains. Furthermore, process models such as waterfall model [10]; components reuse [10], program growth [11], spiral model [12] and evolutionary model [13] have also been developed to guide the execution of activities in software development processes. In general terms, the existing process models present only the common stages of the development processes, without precisely prescribing, for instance, strict order or duration of execution. And most of these models include requirements analysis and specification, design, manufacturing (or implementation – for software products), assembly, inspection 2

followed introductory courses on human and product (in which issues such as the roles of a human both as a consumer and as a user in the development of products and in facilitating the success of a product in the market are addressed), product design I (in which students practice important basic skills of engineering drawing, building prototypes and documenting design concepts); product in action (in which the elementary concepts of engineering statics, such as the concepts of free body diagram and equilibrium of forces and moments in structural members are introduced); and they were following the engineering design course (in which the basics of mechanics of materials, materials science, and industrial production had already been introduced). They were asked to fill in the questionnaire without being told the purpose of the investigation. This was intentionally done to avoid any bias. Furthermore, no formal methods or structured guidelines for execution of activities were introduced to the subjects in advance to avoid any bias or reference to a precedent.

this allows the structural design processes to proceeds intuitively and sometimes in an ad hoc fashion, without following any formal guidelines, formal methods, or strict sequence of execution of the activities. The challenge, therefore, was how to come up with a suitable formal structural design method. The investigation described in this paper focused on addressing this challenge. In the following section, we describe the methodology we used to investigate how structural design processes unfold and to determine the patterns of execution of structural design activities. METHODOLOGY The study was conducted through a structured questionnaire survey among a random sample of 182 students (out of 215 overall), interviews and observations. A short questionnaire that consisted of six closed-ended questions was used as an instrument for gathering data and opinions from the subjects. The idea behind using closed-ended questions was to allow subjects to provide standardized responses that made it easier to compile the collected information. The questionnaire survey gave us some hints regarding the need (or not) for structured procedural process model within the structural design interval, and it also provided some insights into specific aspects such as how the order of execution of activities should possibly be like. The questionnaire survey was complemented by unstructured interviews (i.e., as follow-ups on some of the comments made by some respondents) and observations (i.e., to get a look at what subject actually did in context when performing the structural design activities). These complemented the questionnaire survey and provided further understanding of the issues noted during the analysis of questionnaire survey data, and also helped us to reach more insightful conclusions. One of the immediate tasks before the questionnaire was developed was to identify the activities typically performed in structural design processes. By carefully studying literature about the existing process models such as [4-5], [7-9] and based on our own practical experiences and expert judgments, the tasks listed in Tab. 1 were identified as the principal tasks in the structural design intervals.

Products Three product design assignments on development of three different products were used as case study product development projects in this investigation. The products that had to be designed were: coat hanger, logging truck trailer, and water bike frame. The subjects were asked to come up with design concepts (by performing the activities listed in Tab. 1). As indicated earlier, the subjects were neither demanded to follow any strict order of execution of activities nor to use any specific established process model in this study. The analysis in the context of industrial design engineering structural design process included carrying out product use analysis (investigating how the product will be used and how it will function) and to come up with a list of requirements and wishes. This included identification of the key ergonomics parameters and values (e.g., weights, externally applied forces, dimensions, etc.), which served in the subsequent stages, for instance, as the basis for determination of whether (or not) the conceptualized structural members meet strength or stiffness criteria. Concept development involved: (i) developing design solution principles for parts of the product and selecting viable principles (by using tools such as morphological charts, Harris

Subjects Subjects were carefully selected to ensure that the main purpose of the investigation, which was to correctly gauge sentiments of a population of novice designers and engineers (as far as the practice of structural design is concern) by collecting reliable data, is met. We selected impartial individuals who were not yet fully exposed to or influenced by any practice of using certain specific formal methods. We however ensured that the selected subjects had sufficient expertise that allowed them to undertake structural design tasks in industrial design engineering context. First year industrial design engineering students in their last semester of their first year of study were considered to meet the above-mentioned conditions and were therefore selected to be the subjects in this study. By the time of this investigation, all subjects had already

Table 1. SHORTLISTED OF STRUCTURAL DESIGN TASKS Tasks Use analysis and requirements specification Concepts development Process selection Design presentation (drawing) Evaluation Materials selection Engineering analysis

3

profile2, etc.); and (ii) creating alternative combinations of solution principles (commonly also known as the principal solutions) for a product as a whole and choosing the overall solution principle that meet the previously specified requirements and wishes. Materials selection entailed looking back to requirements and formulating criteria for selecting materials and subsequently using tools such as database queries and search engines to select suitable materials. Process selection involved selecting production process(es) (also by using database queries and search engines), evaluating manufacturability and assemblability of components, and specifying the production process steps. Engineering analysis was the major activity which was also closely tied to the previously mentioned tasks. The subtasks in engineering analysis included investigation of how the envisioned product would be loaded (e.g., would it be subjected to a torsional load?, shear load?, axial load?, etc.), identification of critically loaded parts, sketching free body diagrams of components, sub-assemblies, etc., identification of applicable theories, e.g., for determination of strength and deflection; and using the identified theories to determine stress, deflection and subsequently comparing calculated values with allowable values. Design presentation essentially involved drawing the final design concept and specifying the key dimensions. And evaluation basically entailed looking back to the engineering analysis calculations and to various choices made; and making suggestions for improvement or recommendations for implementation. The subjects formed small design groups (teams) of two or three members and worked together towards achieving the common goal, which principally was to ensure that proper materials for structural members are selected and that the eventual designed structures were strong and stiff enough to withstand different forms of loadings (i.e., shear, torsion, axial and compression loading) apart from meeting other requirements (such as ergonomics, usability, functionality and manufacturability). Each assignment lasted for about 40 hours.

Q3 generated data that we used as the basis for determining if it was useful to provide the list of tasks in advance and if the provided list helped in some ways to streamline the activities. Q1: It was specified prior to the start of the design assignments that you will perform the structural design tasks listed in Tab. 1. Which tasks did you actually do in person? Please indicate. Q2: Did you divide tasks and focused only on one or two tasks? Q3: Did knowing the list of tasks that should be performed in advance help you to streamline the structural design activities? Q4: Did you follow any specific and strict order in the execution of the tasks listed in Tab. 1? Q5: Which task did you perform first, second, etc. Please indicate. Q6: Do you think it was important to indicate the order of execution of the tasks listed in Tab. 1?

DATA ANALYSIS AND RESULTS The data collected from the questionnaire survey, interviews and observations were analyzed in order to extract as much information as possible and to draw meaningful conclusions on the conducts of structural design processes. This study gave us new insights into how structural design in industrial design engineering processes unfolds, revealed how subjects actually pursued the tasks and helped us to uncover if subjects followed any patterns or executed tasks in any specific order. Q5 was specially designed to give us insights into the patterns and the order by which the subjects performed the tasks listed in Tab. 1. It required subjects to indicate the order in which they executed the structural design tasks. Based on the collected Q5 data, the chronological position preference ratio, (t, c) for each task – chronological position combination was then determined by using the equation below. m

Data Collection A short questionnaire that consisted of six closed questions listed below and labeled Q1 through to Q6 was used to collect data and opinions of subjects. Subjects were asked to respond to all questions. The collected data was analyzed to determine the statistical trends and to uncover if there were any notable patterns in the execution of the tasks listed in Tab. 1. The data collected from responses to Q5 were analyzed to determine the patterns or the order by which the subjects performed the tasks while data collected from responses to Q1 and Q2 were analyzed to understand the manner by which the subjects conducted the structural design tasks. Furthermore, Q4 and Q6 data helped to gauge the feelings of the subjects on whether or not it was important to specify in advance a strict order and/or guidelines for execution of structural design activities, while

 (t , c ) 

x i 1 n

y j 1

i

(1)

j

where n is the total number of subjects who responded to the question item; m is the number of subjects who responded positively to the item (i.e., by indicating that they executed the tasks at a particular chronological position), t is the task, and c = 1, 2, …., 7 is the chronological position. x = 1, is the respondent’s positive response count and y = 1 is the total (i.e., positive and negative) response count. The obtained chronological position preference ratios (t, c) are summarized in Fig. 13. This summary helped us to visualize the chronological order preferences and thus the 3

As an example, the value in the first cell in Fig. 1 (i.e., 0.904) was obtained by substituting x=1 and y=1 in Equation 1; while 165 subjects (m =165) responded positively to the question item and the total of subjects who responded to the question item was 182 (n=182).

2 Graphic representation of the strengths and weaknesses of design concepts (see e.g., http://wikid.eu/index.php/Harris_profile)

4

chronological position, c

Use analysis and requirements specification

Concepts development

Process selection

Design presentation (Drawing )

Evaluation

Materials selection

Engineering analysis

Chronological Position

0.904

0.059

0.011

0.000

0.000

0.000

0.027

1

0.044

0.852

0.022

0.016

0.000

0.022

0.044

2

0.010

0.047

0.193

0.307

0.010

0.188

0.245

3

0.005

0.005

0.202

0.074

0.016

0.431

0.266

4

0.000

0.006

0.349

0.077

0.047

0.278

0.243

5

0.000

0.013

0.208

0.384

0.170

0.063

0.164

6

0.000

0.000

0.006

0.131

0.844

0.013

0.006

7

tasks, t

Figure 1. A (t,c) DIAGRAM SHOWING PATTERNS OF PERFORMING STRUCTURAL DESIGN ACTIVITIES – THE HIGHEST CHRONOLOGICAL POSITION PREFERENCE RATIO  VALUE(S) FOR EACH STRUCTURAL DESIGN ACTIVITY IS HIGHLIGHTED IN BLACK. shown in Fig. 2 (and formally structured as shown in Fig. 3) was derived. Figure 4 shows further analysis results. The majority of the respondents (85.4%) felt that there was a need to specify formal chronological order of executing activities whilst 14.6% did not feel this way, as shown in Fig. 4 (a). Furthermore, it is apparent from this analysis that although the subjects were supposed to work in groups of two or three members, 48.6% of them did not really worked in groups, but instead divided the tasks, and as a result of this, each subject focused only on some selected activities; 73.9% of the subjects felt that simply knowing the list of tasks that should be performed in advance was already helpful enough for streamlining the structural design activities while 26.1% did not feel this way, and 72.2% of the subjects indicated that they followed certain specific and strict order in executing the tasks listed in Tab. 1 while 27.8% did not – see Fig. 4 (b), (c), and (d). This suggests that the subjects intuitively decided on their own on the best way to structure the structural design process and strictly adhered to a certain preferred order of execution of tasks. Although the obtained results seem clearly to point to certain specific trends and patterns, further studies still need to be carried out to verify the validity of the obtained results.

patterns by which the subjects performed the structural design tasks listed in Tab. 1. Based on the obtained chronological position preference ratio () values, it can be seen that: (1) use analysis and requirements specification and (2) concept development and selection tasks were the first and the second activities respectively in the order of execution while evaluation was the last executed activity for most of the subjects. It appeared that the first two tasks (i.e., use analysis and requirements specification – with  = 0.904 and concepts development - with  = 0.852) were predominantly executed consecutively in the early stages of the structural design process while evaluation was the final activity for most of the subjects (with  = 0.844). The obtained  values (see Fig. 1) further suggest that engineering analysis was closely tied to materials selection and process selection; and these activities were executed interchangeably as the 3rd, 4th, or 5th task in the order of execution (see Fig. 2). In other words, the  values suggest that process selection and materials selection were mostly performed recursively alongside engineering analysis. This suggests that these three activities (i.e., process selection, materials selection and engineering analysis) should actually be performed after concepts generation in any order of execution. For most subjects, formal representation of the design concepts (i.e., design representation) also started after concepts generation (first by sketching visual representations of the design concepts and showing the assumed main dimensions), then ceased to allow for the execution of engineering design, materials selection, process selection tasks, and establishment of the main dimensions, and was finalized after completion of these three recursive tasks. Based on the obtained  values, a chronological order for execution of structural design activities

DERIVED SCHEMATIC PROCEDURAL MODEL FOR GUIDING THE EXECUTION OF STRUCTURAL DESIGN TASKS The study gave us some indications on how structural design processes should be conducted in practice. The analysis results suggest that the subjects intuitively followed the order of execution of tasks depicted in Fig. 2 and elaborated in Fig. 3. Figure 2 shows the pattern of execution of the structural design 5

and is important in terms of the organization of activities and quality improvement – see e.g., [15]. As pointed out earlier, most of the existing methods are, however, coarsely defined and only guide the execution of high level activities within the product development interval. Also, some activities and relationships among activities within the product development interval and even within sub processes such as the design process are rather complex, and we therefore argue in this work that it would be beneficial to come up with a dedicated formal procedural process model for guiding the activities within the sub processes. The objective of this study was therefore to explore how the structural design process progresses in practice and to determine if there is a real need for specific formal methods for organizing and guiding structural design activities. We specifically focused on structural design in industrial design engineering practice and on the products traditionally designed by industrial design engineers. A study which involved a questionnaire survey, interviews, and observations was conducted and has led to formulation of the structured linear – cyclic procedural process model shown in Fig. 4. The execution of the structural design activities by using the proposed procedural process model is rather straightforward. The design process has to simply pass through the process steps specified in Fig. 4 and described in the previous Section. It is widely known that quality of products largely depends on quality of the engineering processes through which they are developed - see e.g., [3], [15], [16]. Hence, we argue that following the specified process steps and consistently working according to the framework of the proposed linear – cyclic procedural process model can have a huge impact on the quality of the eventual product. During the execution, if something does not make sense at any sub-phase, the designers and engineers obviously need to refine their in-process deliverables right at that particular sub-phase and should also make sure that any changes made are reflected and accommodated in other subphases as well.

tasks intuitively followed by the subjects, which we used as the basis for generalization and for creating the linear-cyclic schematic procedural model for the execution of structural design tasks shown in Fig. 4. Since it stems from the data and the information gathered from real structural design processes, we argue that it can be used in many structural design processes to help designers and engineers to organize and execute structural design tasks. It is linear and cyclic in nature in that it splits the structural design process into multiple phases and suggests both sequential and cyclic flows of tasks. This allows users to perform structural design tasks both linearly and recursively. This systematic structural design procedural process model does not require users to follow a formal structured order in the execution of the mid-way tasks that follow after concepts generation. It demands that engineering analysis, process selection, materials selection and design presentation tasks should navigate randomly and intuitively until an optimal design concept is achieved. This is to say that the execution of these tasks should not necessarily be in a certain strict sequence. And the decisions regarding the order of execution to follow should ideally be left entirely to the designers and engineers involved in the process. Despite the clear evidence from this investigation that the structural design tasks were executed in an orderly fashion and with a clear pattern, some subjects, however, still expressed mixed opinions and were somewhat skeptical on whether or not it is indeed important to have a formal chronological order for execution of structural design tasks - refer to Fig. 4(c). Some of the subjects involved in the investigation felt strongly that it was only sufficient to know the identity of the tasks. DISCUSSION Many formal methods for guiding the design process have been proposed by various engineering design researchers in the past decades - see e.g., [4-5], [7-9]. It has long been argued in many publications that following formal methods in the execution of product development activities is the best practice

1

Use analysis and requirements specification ( = 0.904)

2

Concepts development ( = 0.852)

3

4

5

6

Materials selection Process selection

Process selection

Materials selection

Process selection

Engineering analysis

Engineering analysis

Engineering analysis

Design presentation (Drawing)

Material selection

Process selection

7

Design Evaluation ( = presentation 0.844) (Drawing)

Figure 2. DERIVED CHRONOLOGICAL ORDER FOR THE EXECUTION OF STRUCTURAL DESIGN ACTIVITIES 6

Engineering  analysis Analysis &  requirements  specification

Design  presentation  (drawing) 

Concepts  development

Evaluation  Process  selection

Materials  selection

Figure 3. LINEAR-CYCLIC PROCESS STEPS OF EXECUTION OF STRUCTURAL DESIGN TASKS – DERIVED FROM THE WAY IN WHICH DESIGN TRAINEES PERCEIVE AND EXECUTE THE STRUCTURAL DESIGN ACTIVITIES, SET BASED ON THE (c,t) DIAGRAM (SEE ALSO FIG. 1) Also, since an in-process deliverable (i.e., a given form of design documentation or any form of generated product data) is produced at every stage of the structural design process, it is easier to look back and evaluate if the previously formulated requirements are met as the design process progresses. The fact that the model is partly linear can, however, also be a disadvantage of this procedural process model because going back a step if something within the structural design sub-phase has gone wrong can be very complicated. Also, the designers and engineers are not often very clear of the exact expectations, and changes may always emerge in between the linearly executed activities; and this may cause a lot of confusion. For instance, minor changes or errors that arise e.g., in the calculations may cause a lot of problems, e.g., may lead to selection of another material, consideration of another manufacturing process, etc. Another major limitation is that this procedural process model allows the involvement of the stakeholders (i.e., customers, end-users, etc.) only in the beginning of the structural design process. For instance, customers cannot express their opinion after the analysis phase, e.g., on if what is being designed is exactly what they had asked for. This is possible only after completion of the structural design cycle, when the final drawings (e.g., 3D CAD models or orthographic computer generated models) and other design documentations are ready. In conclusion, despite the above mentioned shortcomings, we argue that it is important to adhere to the steps specified in the proposed linear – cyclic structural design procedural model. According to the findings of the investigation, the majority (~ 70%) of the subjects of the investigation intuitively followed the depicted linear-cyclic sequence and this helped them to attain the expected quality of the in-process deliverables as well as of the eventual product concepts. The end users of the product always expect high quality, low cost, and timely delivery of products. To this end, further studies are needed to investigate if the proposed procedural process model can contribute to achieving high quality product at a low cost and on time. It should also be noted that apart from using formal procedural models; other practices such as capturing and representation of vital knowledge and using it in various stages

The proposed process model is nothing but documented steps that guide the designers and engineers to repeat the same work repeatedly with the same precision and predictability, and provides a consistent way for structural design of products. One of the advantages of this procedural process model for structural design is that despite being partly a linear and partly a cyclic procedural model, it is still very simple to implement. 14,6

51,4 48,6

Yes

85,4

No

(b)

(a) 26,1 27,8

72,2

73,9

(c)

(d)

Figure 4. RESULTS OF THE ANALYSIS OF THE DATA COLLECTED IN RESPONSE TO THE FOLLOWING QUESTIONS: (A) DO YOU THINK IT WAS IMPORTANT TO INDICATE THE ORDER OF EXECUTION OF THE ABOVE LISTED IN TAB. 1? (B) DID YOU DIVIDE TASKS AND FOCUSED ONLY IN ONE OR TWO ACTIVITIES? (C) DID KNOWING THE LIST OF TASKS THAT SHOULD BE PERFORMED IN ADVANCE HELP YOU TO STREAMLINE THE STRUCTURAL DESIGN ACTIVITIES? (D) DID YOU FOLLOW ANY SPECIFIC AND STRICT ORDER IN THE EXECUTION OF THE TASKS LISTED ABOVE? 7

of the product life also have impact, e.g., in terms of cost, sustainability and energy use [17]. SUMMARY, CONCLUSIONS AND FUTURE WORK In the work presented in this paper, we explored how structural design in industrial design engineering environment unfolds through non-structured observations, interviews, and a questionnaire survey. A number of interesting observations have emerged from this study and worth to be considered in managing and coordinating structural design tasks. The feeling of the overwhelming number of subjects was that there is a need for formal procedural model for guiding the structural design process. On one hand, there was a general consensus among the subjects on the need of having in place tightly defined order of execution of tasks and on the necessity of following formal procedural process steps and guidelines in the beginning of the structural design process, namely during analysis and concepts generation. On the other hand, the investigation results suggest that the order of execution of the mid-way activities of materials selection, process selection and engineering analysis is unimportant. Overall, we argue that a compromise approach would be to use a loosely defined (i.e., in terms of the sequence of activities) linear – cyclic procedural process model introduced in this paper, which accommodates both of the above described wishes of the subjects. It is important to emphasize that since the study involved only a handful of selected structural design processes, we cannot draw definitive conclusions regarding the order or the process steps of the structural design process. Therefore, further studies are needed to verify and validate the obtained results. Furthermore, it would also be worthwhile to study the applicability of the proposed linear-cyclic procedural process model in real world and to determine how it could possibly be used alongside the traditional design process models such as those proposed in [4-5], [7-9]. It should also be noted that although most designers followed the observed order of execution of tasks, this is not necessarily a “good” process. Further studies are required to correlate the observed process model and order of execution of tasks with the objective design performance. In addition, formal design methods usually prescribe some general principles (e.g., the two axioms in [9]) that designers should follow. This is this is also one of the issues for future research.

[2].

Opiyo, E. Z., Horváth, Vergeest, J. S. M., 2002. “Quality Assurance of Design Support Software: Review of the state of the art”. Computers in Industry, Vol. 49 No 2. pp. 195-215.

[4].

Pahl, G. & Beitz, W., 1993. Engineering Design. translated by K. Wallace (Design Konstruktionslehre Methoden und Anwendug), 3rd Edition, Springer-Verlag, Berlin.

[5].

Roozenburg, N. F. M. and Eekels, T., 1995. “Product Design: Fundamentals and Methods”. Wiley Series on Production Development: Planning, Designing, Engineering, John Wiley & Sons.

[6].

Weustink, I.F. ten Brinke, E. Streppel, A.H., Kals H. J. J., 2000. ‘A generic framework for cost estimation and cost control in product design’. J. of Materials Processing Technology 103, 141-148.

[7].

Blessing, L. T. M., 1994. “A Process-Based Approach to Computer-Supported Engineering Design”. Ph.D. Thesis, University of Twente, Enschede, The Netherlands.

[8].

Andreason, M. M. and Hein, L., 1987. “Integrated Product Development”. IFS (Publications), Ltd/Springerverlag, Bedford, UK.

[9].

Suh, N. P., 1995. “Axiomatic design of mechanical systems”. Transactions of the ASME, A Special Combined Issue of the J. of Mechanical Design and the J. of Vibration and Acoustics 117: 2-10.

[10]. Jones, W. G., 1990. Software Engineering. John Wiley and Sons. [11]. Brooks, F. P., 1987. "No Silver Bullet: Essence and Accidents of Software Engineering". Computer, No. 20. Vol. 4 pp. 10-19. [12]. Boehm, B. W., 1988. "A Spiral Model of Software Development and Enhancement". Computer, No. 31. Vol. 5 pp. 61-72. [13]. Lehman, M. M., 1980. "Programs, Life Cycles, and Laws of Software Evolution". Proc. of the IEEE, Vol. 68 No. 9, pp. 1060-1076. [14]. Pfleeger, S. L., and Hatton, L., 1997. "Investigating the Influence of Formal Methods". Computer, vol. 30, no. 2, pp. 33-43, Feb. 1997, doi:10.1109/2.566148.

REFERENCES [1].

[3].

[15]. Opiyo, E. Z., (2003) “Facilitating the process of development of design support systems”, Ph.D. Thesis, Delft University of Technology.

Paulk, Mark C.; Weber, Charles V.; Garcia, Suzanne M.; Chrissis, Mary Beth C.; and Bush, M., 1993. "Key Practices of the Capability Maturity Model Version 1.1". Software Engineering Institute. Paper 171. http://repository.cmu.edu/sei/171

[16]. Beizer, B., 2001. "Software Is Different". Quality Techniques Newsletter, Software Research Inc., http://www.soft.com/News/QTN-Online , June Issue.

Langari, Z., and Pidduck, A. B., 2005. 'Quality, Cleanroom and Formal Methods'. Proc. of 3-WoSQ ‘05, May 17, 2005, St Louis, Missouri, USA/ ACM SIGSOFT Software Engineering Notes Homepage, Vol. 30 No. 4, July 2005, pp. 1 - 5.

[17]. Chandrasegaran, S. K., Ramani. K., Sriram, R. D., Horvath, I., Bernard, A., Harik, R. F., Gao, W. (2013) The evolution, challenges, and future of knowledge representation in product design systems, ComputerAided Design Vol. 45, No 2, pp. 204-228 8