Journal Search Engine
Search Advanced Search Adode Reader(link)
Download PDF Export Citaion korean bibliography PMC previewer
ISSN : 1598-7248 (Print)
ISSN : 2234-6473 (Online)
Industrial Engineering & Management Systems Vol.16 No.4 pp.465-478

Designing Reliability Improvement Model Using Design Structure Matrix

Mehdi Karbasian*, Sayed Mohammad Kazemi, Bijan Khayambashi, Sayed Akbar Nilipour
Industrial Engineering Department, Malek-Ashtar University of Technology, Isfahan, Iran
Corresponding Author,
20170728 20170805 20170821


As an inherent characteristic of any product or system, reliability is one of the parameters of design, construction and operation to be considered and controlled as an important criterion during relevant processes. The need to ensure the continuous and proper functioning of a product has led designers to pay special attention to increasing reliability. Therefore by coordinating all design experts and support groups throughout the implementation of product life cycle, the product development life cycle period can be reduced and value-creation of products increased. As a result, the existence of a comprehensive model is of vital importance. In this study, all industrial engineering techniques effective in improving reliability are first identified and then prioritized in all phases of the product life cycle using multicriteria decision making method. Finally, using design structure matrix, priority and posteriority of all techniques is determined in each phase and a systematic model is provided to improve equipment reliability.



    In recent years, the accelerated development of science and technology and the emergence of new fields of science and technology have resulted in designing and building highly sensitive and sophisticated systems commonly applicable in industries requiring high reliability. In these systems, malfunction of sensitive parts may cause irreparable damage. The need to ensure the continuous and proper functioning of these products has thus led designers to pay special consideration to increasing reliability. Various methods and tools need to be used to improve system reliability and a specific level of reliability is needed for each system depending on performance type and defined requirements. In recent years, a large number of management and industrial engineering tools and techniques have been created. In addition to management science and industrial engineering, mechanical, electronic and material engineers have also devised and introduced a lot of reliability improvement methods to production and service sectors; however, the ambiguity and confusion of different organizational levels in employing these techniques have posed great challenges for the management.

    For successful selection and use of these tools, managers must be aware of these techniques and their priority and posteriority. With the advent of these tools, there are also reports indicating under what conditions each management tool and technique will bring the best results. Most available researches refer to the strengths of the tool without having a holistic and all-tools-together view. There is also no significant report on evaluating these tools. Without adequate studies, selection and use of management tools will become a dangerous and risky game. More importantly, in some cases, these techniques have caused confusion in some industries leading to pessimistic views toward consultants and academics, merely considering the use of these techniques as undesirable cost.

    On the other hand, it is quite essential to have a detailed plan to document the duties, procedures, tools, analyses and tests required for a given system to achieve the desired level of reliability from the beginning of system design and development. For better understanding of all partners effective in designing prototype and mass production, a model is needed so that all relevant people can easily and efficiently accomplish their tasks toward achieving desired reliability without getting puzzled by those techniques, methods and operation The existence of a road map and model on the basis of which product reliability can be improved is a necessity these days. Having examined previous studies, it is found that current models available to the country’s industries and organizations are too general to be suitably used for reliability improvement in all industries. Therefore, combining all available reliability improvement techniques for product life cycle, multi-criteria decision making method and design structure matrix are used in this paper to introduce a comprehensive model to improve reliability.


    The word ‘reliability’ developed from the word ‘rely’, which is defined as a ‘sense of dependence or trust and perhaps has a notion to fall back on’ (Online Etymology Dictionary, 2014). It was first used as early as 1816 by the poet Samuel T. Coleridge, who wrote about his friend who inspired everybody around him with “perfect consistency and absolute reliability” (Saleh and Marais, 2001). Since then the concept of reliability has become rather popular, and is used extensively by the general public as well by the technical community (Conradie et al., 2015).

    Given the increasing concern for quality and high reliability, extensive research has been done on this issue. For example, Onoufriou makes an assessment over reliability in structural, steel and marine structures under severe environmental load (Onoufriou and Forbes, 2001). Avontuur studies the role of reliability in design process of train movement system and introduced a new method to analyze reliability of mechanical and hydraulic systems (Avontuur and van der Werff, 2001; Avontuur and van der Werff, 2002). Zou et al. (2002) studies car door reliability taking into account the amount of energy used to close the door as an important quality factor and also examines reliability in a random flow network considering the probability of failure in nodes and arcs (Lin, 2002). Dutuit and Rauzy (2005) estimated reliability using fault tree analysis by binary decision matrix in a system with independent reparable components.

    With a different perspective, Ascher (2010) investigated reliability improvement of components and systems using Decreasing Force of Mortality technique, Doguc and Ramirez-Marquez (2009) introduces a probabilistic model for the application of Bayesian networks in calculating reliability, and Der Kiureghian and Song, (2008) analyzed multi-scale reliability and studies updating complex systems using linear programming.

    Wilson et al. (2011) studies reliability under uncertainty using Bayesian approach in two series and parallel systems and Bichon analyzes system reliability with multiple failure modes using alternative Gaussian model. Aubert calculates life reliability of triode lamp isolated package (Bichon et al., 2011; Aubert et al., 2011) and Lee calculates reliability of systems with finite elements with the effect of fatigue caused by successive failures using branch and bound method (Lee and Song, 2012).

    Castellazzi et al. (2011) discussed reliability improvement of power generation system using dynamic coolers, Dong et al. (2012) predicts reliability and fatigue of pipe joints welding in the supporting structure of a fixed jacket of an offshore wind turbine at a depth of 70 meters. Frémont et al. (2012) offered a model to predict reliability of microelectronics sets, Sankaraiah et al. (2011) attempted to estimate influence of numerous limitations on system reliability, and Conrad studies reliability failure rate in railways. All above studies clearly refer to only a limited area of reliability – none of them consider the overall process of product design to achieve the desired reliability of systems and subsystems.


    2.1.Models Included in the Calculation and Improve Reliability

    2.2.1.NASA Company Model

    The model depicted in Figure 1 presents the monitoring and evaluation of management processes in an ordered way, as a general scheme by NASA Company and it has been considered theoretically, also mentioned this point that more details should be investigated in fact. This General scheme deals with the detection of faults caused by tests as well as the reliability assignment and planning (AMSC and SESS, 2011).

    2.2.2.Lee Model

    Lee et al. (1991) showed eight initial phases in order to achieve reliability as shown in Figure 2. This Figure shows eight initial phases in designing effective reliability. These phases include mission definition, design guidelines, analysis design, parts reliability, periodic reviews design, manufacturing, assessment testing and operational testing. In the first phase, product mission is defined and in the second phase, necessary reliability and costs are considered. In the third phase, design engineers propose additional design instructions such as necessary load criteria, design margins and necessary standards of parts. In design analysis, the third phase, thermal stress, electrical cases, tolerances and scheduling are investigated and fault and failure probabilities are analyzed in order to predict maintainability and reliability of the system. This prediction is conducted in order to compare existing system with the ideal one. In the next phase, parts reliability is extensively investigated, suppliers and reliability requirements are considered, and non-standard parts that can’t provide the desired capability are identified and modified. In the fifth phase, design is investigated periodically. In the sixth phase, operational tests and normal growth are conducted by a series of tests to estimate and compare real system with ideal one. In the seventh phase, tests are evaluated and in the last phase, the development of process productivity and training programs for product are considered (Lee et al., 1991).

    2.2.3.Safie Model

    Safie et al. (2012) employees of spatial company of NASA, showed reliability in the phases of product design as shown in Figure 3. This model divides reliability into reliability design and reliability process. In reliability design, tension and strength are considered as two dependent variables. Tension variable is considered as a function of environment, loads and operational condition. The process critical parameters, processing method and process control are considered in process discussion.

    2.2.4.Levin Model

    Levin and Kalal (2003) divided reliability process into some phases including conceptual design, production and product life end in their book as shown in Figure 4. In conceptual design phase, product concept is defined based on business and market requirements and then functions and design adequacy are confirmed in design phase. In production phase, design is transformed into product and necessary optimization is conducted and finally, worn-out parts and materials are removed from warehouse inventory and new products are replaced after the useful life end of old product.

    2.3.Determination of Product Life Cycle Phases Using Existing Standards

    Product life cycle covers all necessary activities from designing product and procurement raw materials to delivering finished product and offering post-sale service. These activities include research and development (R&D), product designing, manufacturing, sale, marketing, advertising and post-sale service. Following points were investigated in order to determine necessary phases for the completion of product life cycle:

    • 2.3.1. Product life cycle covers all necessary activities from designing product and procurement raw materials to delivering finished product and offering post-sale service. These activities include research and development, designing product, manufacturing, sales, marketing, advertising and post-sale service. Figure 5 shows the product life cycle system proposed by Blanchard and Fabrycky (2010).

    • 2.3.2. According to Figure 6, shows product life cycle in the organization of aero-space industries (Blythe et al., 2010).

    • 2.3.3. NASA Company proposed eight phases in 7120 standard as follows below and according to the completeness of the model relative to other models, it was considered as a desirable model in this research. Eight phases include:

    • 1. Requirement analysis

    • 2. Feasibility and conceptual design

    • 3. Preliminary design

    • 4. Final design

    • 5. Manufacturing Prototype and engineering development

    • 6. Product design and development

    • 7. Process design and development

    • 8. Supporting operations application and confirmation

    2.4.A Model for Improving Reliability

    There are many instruments and methods with their own advantages for design engineers that are applicable for different phases of designing and product manufacturing. Hence, in this research, we have tried to identify effective techniques on product life cycle and classified them in each phase so that they can be used optimally. The study was conducted in an industrial plant in Esfahan, Iran, and 28 people from design department completed a number of questionnaires uninterruptedly.

    2.4.1.Flowchart for Reliability Improvement Model

    To the best of our knowledge, there is no published document pertaining reliability improvement model in industries and organizations in Iran. As a result, after literature review, reliability improvement models were gathered using various English documents, articles and books and then some criteria and techniques affecting reliability were extracted through interviews with experts in industry and academia. Finally, taking into account all requirements and conditions of the mentioned industry, the appropriate reliability improvement model is presented.

    As evident, all previous studies refer to a limited area of reliability process – none of them consider the overall process of product design to achieve the desired reliability of systems and subsystems. Hence, as depicted in conceptual model in Figure 7, this research attempts to classify all techniques for eight phases of product life cycle.

    2.4.2.Identifying Necessary Techniques for Reliability Calculation and Improvement in All Phase of Product Life Cycle

    There are many instruments and methods with their own advantages for design engineers that are applicable for different phases of designing and product manufacturing. Hence, in this research, we have tried to identify effective techniques on product life cycle and classified them in each phase so that they can be used optimally.

    The study was conducted in an industrial plant in Esfahan, Iran, and 28 people from design department completed a number of questionnaires uninterruptedly. A series of techniques were selected from industry engineering handbooks and resources including 68 defining techniques, with the assistance of experts in the design and engineering departments of Armaments Industries in Isfahan province. These techniques are shown in Table 1.


    3.1.Scoring Criteria of any of the Methods

    Assigning each technique to each phase needs scoring and determining priorities. Therefore, a meeting was held with the participation of all industry experts. At first, all criteria were determined through brainstorming. These criteria could be used to rank techniques. After that, these criteria were classified in priority using pairing comparisons technique and four criteria were selected as the most important criteria in using techniques in each phase. The weight of each criterion was determined using ANP method as shown in Table 2. The determined criteria include:

    • a. Cost of technique performance

    • b. Required time of technique performance

    • c. Application extent in each standard phase

    • d. Simplicity of technique performance

    3.2.Assigning Each Technique to one or More Phases of the Cycle

    One of these methods is “similarity to ideal solution” presented in 1981 by Yoon & Hwang and welcomed by researchers and various users. In this method, options are ranked based on similarity to an ideal solution such that a more similar option to the ideal solution will gain a higher rank. With strong mathematical background, this decisionmaking method uses two concepts of “ideal solution” and “similarity to ideal solution.” Ideal solution, as its name implies, is the optimum solution which may not exist practically but we try to approach.

    Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) is a useful method to deal with multi-criteria decision making in real world (Hwang, 1981). Using ideal point concept, it is easy to understand TOPSIS which is why we employed this method in this paper.

    In order to allocate any of the techniques to previously determined phases, “similarity to ideal solution” was used as a multi-criteria decision making method and all relevant industry experts were questioned using 8 questionnaires in each phase on the basis of above criteria. Having collected all answers, the results were analyzed in Excel and each technique was assigned to desired phases according to Table 3.

    3.3.Determining Priority/Posteriority Relationship of Techniques in Each Phase using Design Structure Matrix (DSM)

    DSM is a modeling tool in system engineering used for development of complex systems and enables analysis of model and the dependencies between elements of the system to understand system behavior in order to improve the overall functionality of the system.

    Design structure matrix is a powerful modeling tool allowing users to decompose a system into various subsystems in order to have awareness about the strengths of integrating elements of a system. It also provides insight for a complex design structure to formulate clustering for the expansion of modules (Shamsuzzoha and Helo, 2012).

    DSM was first introduced by Prof Steward at California State University in the 1970s in a study titled “Design Structure Matrix: a way to design complex systems.” In this study, this matrix was used as a tool to identify dependencies between tasks and to sequence the development process (Carrascosa et al., 1989).

    Steward developed concepts like principal-circuits, shunts, partitioning, and tearing as well as PMS32 software tool for modeling and manipulation of DSMs (Jin and Huang, 2010). By the mid-1990s, DSM was not used in industrial applications until some professors and graduate students at MIT applied it in various military and non-military industries.

    After determining the priority of each phase, the relationships between techniques were examined in each phase. In order to further explore relationships, DSM was used.

    To determine the relationship between techniques in each phase, a questionnaire was used to identify relationships between components of each phase (Table 6) and relationships were then analyzed using PSM32 (Table 7 and Table 8) which led to following results:

    3.3.1.The First Phase of Requirements Analysis

    As seen, development of human resources and staff training about reliability of a given product, explanation of reliability demanded by customer, and pattern making about reliability of similar products were all placed in the first level. Techniques for identifying assumptions, limitations, goals, and requirements of product reliability were placed in separate levels. In requirements analysis phase, information in Table 4 were analyzed by the software in accordance with Table 5 results of which are shown in Table 6.

    Using above levelling, in Figure 8, known as ‘DSM developed model in first phase,’ first-level criteria 2, 3, and 4 lie on the first level of Figure and other techniques lie on other levels accordingly. To sum up, other tables are not presented in this paper.Figure 9Figure 10Figure 11Figure 12Figure 13Figure 14Figure 15

    3.3.2.Phase II: Design Concept

    3.3.3.Phase III: Preliminary Design

    3.3.4.Phase IV: The Final Design

    3.3.5.Phase V: Development

    3.3.6.Phase VI: Design and Product Development

    3.3.7.Phase VII: The Design and Development Process

    3.3.8.Phase VIII: Implementation and Verification

    3.4.Validation of Reliability Improvement Model

    To validate the model, a questionnaire answered by industry experts was used for analysis. The results in Table 7 show that all respondents expressed their agreement with the following variables with an average score of greater than 4: 1) agreement with the eight-phase model, 2) model operability, 3) compatibility with the main purpose of research, and 4) the life cycle used in model implementation.

    Based on Table 8, all respondents with an average of more than 2 confirmed the posteriority/priority of the 8-phase model.

    According to Table 9, all respondents with an average of more than two confirmed the number of levels related to the eight-phase model parts.

    As Table 10 indicates, all respondents with an average of more than two confirmed the number of phases and levels of model part.


    In this research, after identifying existing cycles in product life area, all techniques that can be effective in reliability improvement were determined. Then, previous models in the area of reliability improvement were investigated. After selecting more complete life cycle and similarity to ideal option technique, one of the multicriteria decision techniques, all identified techniques were classified between product life phases. Also, priority and posteriority of all techniques is determined by using design structure matrix and for the first time, in order to understand and develop relationship between equipment’s reliabilities, in the direction of equipment’s reliability, a systematic model was obtained. In other words, combining all available reliability improvement techniques for product life cycle, multi-criteria decision making method and design structure matrix are used in this paper to introduce a comprehensive model to improve reliability. A more comprehensive model should be used in future research compared with existing model and necessary human, time and financial resources related to it should be considered. Therefore, integrated models with project management can be used. Existing techniques in project management help reliability improvement in systems and subsystems and it is possible to achieve the main purpose that is techniques implementation, in industry practically. Results revealed the effectiveness of proposed method in designing reliability improvement model.



    The reliable growth management model MIL-HDBK-189C (2011).


    Lee et al. model in 1991.


    Safie et al. model in 2012.


    Levin (2003) Method.


    Fabriki and Blanchard model (2010).


    Technical review in product life cycle.


    Research conceptual model.


    Basic developed model of DSM in 1st phase.


    Basic developed model of DSM in 2nd phase.


    Basic developed model of DSM in 3rd phase.


    Basic developed model of DSM in 4th phase.


    Basic developed model of DSM in 5th phase.


    Basic developed model of DSM in 6th phase.


    Basic developed model OD DSM in the 7th phase.


    Basic developed model of DSM in the 8th phase.


    List of techniques

    Measures weight defined by ANP

    Technique’s score

    Access matrix for requirements analysis phase

    Software input information in the first phase

    Output results of PSM 32 software in the first phase

    Average data on general questions in model validation section with hypothetical average 3

    Average data on model posteriority/priority with hypothetical average 2

    Average data on the number of model levels with hypothetical average 2

    Average data on the number of phases and model levels with hypothetical average 2


    1. AMSCAMSC (2011)
    2. AscherH. (2007) Different insights for improving part and system reliability obtained from exactly same DFOM ?ofailure numbers ??. , Reliab. Eng. Syst. Saf., Vol.92 (5) ; pp.552-559
    3. AubertA. JacquesS. PA(c)tremontS. LabatN. FrA(c)montH. (2011) Experimental power cycling on insulated TRIAC package: Reliability interpretation thanks to an innovative failure analysis flow. , Microelectron. Reliab., Vol.51 (9-11) ; pp.1845-1849
    4. AvontuurG.C. van der WerffK. (2001) An implementation of reliability analysis in the conceptual design\ phase of drive trains. , Reliab. Eng. Syst. Saf., Vol.73 (2) ; pp.155-165
    5. AvontuurG.C. van der WerffK. (2002) Systems reliability analysis of mechanical and hydraulic drive systems. , Reliab. Eng. Syst. Saf., Vol.77 (2) ; pp.121-130
    6. BichonB.J. McFarlandJ.M. MahadevanS. (2011) Efficient surrogate models for reliability analysis of systems with multiple failure modes. , Reliab. Eng. Syst. Saf., Vol.96 (10) ; pp.1386-1395
    7. BlanchardB. FabryckyW. (2010) Systems Engineering and Analysis., Prentice Hall,
    8. BlytheM.P. SaundersM.P. PyeD.B. VossL.D. MorelandR.J. SymonsK.E. BromleyL.K. (2014) NASA Space Flight Program and Project Management Handbook,
    9. CarrascosaM. EppingerS.D. WhitneyD.E. (1989) Using the design structure matrix to estimate product development time , Proceedings of the ASME Design Engineering Technical Conferences (DETC), ; pp.1-10
    10. CastellazziA. ChoyW.J. ZanchettaP. (2011) Dynamic active cooling for improved power system reliability. , Microelectron. Reliab., Vol.51 (9-11) ; pp.1964-1967
    11. ConradieP.D. FourieC.J. VlokP.J. TreurnichtN.F. (2015) Quantifying system reliability in rail transportation in an ageing fleet environment. , S. Afr. J. Ind. Eng., Vol.26 (2) ; pp.128-142
    12. Der KiureghianA. SongJ. (2008) Multi-scale reliability analysis and updating of complex systems by use of linear programming. , Reliab. Eng. Syst. Saf., Vol.93 (2) ; pp.288-297
    13. DogucO. Ramirez-MarquezJ.E. (2009) A generic method for estimating system reliability using Bayesian networks. , Reliab. Eng. Syst. Saf., Vol.94 (2) ; pp.542-550
    14. DongW. MoanT. GaoZ. (2012) Fatigue reliability analysis of the jacket support structure for offshore wind turbine considering the effect of corrosion and inspection. , Reliab. Eng. Syst. Saf., Vol.106 ; pp.11-27
    15. DutuitY. RauzyA. (2005) Approximate estimation of system reliability via fault trees. , Reliab. Eng. Syst. Saf., Vol.87 (2) ; pp.163-172
    16. FrA(c)montH. DuchampG. GraciaA. VerdierF. (2012) A methodological approach for predictive reliability: Practical case studies. , Microelectron. Reliab., Vol.52 (12) ; pp.3035-3042
    17. HwangK. (1981) Similarity to Ideal Solution (TOPSIS) as a useful method to deal with multi-criteria decision making in real world., Springer-Verlag,
    18. JinS. HuangS. (2010) Quality assessment planning using design structure matrix and resource constraint analysis, M.S. mechanical Engineering., Graduate School of the University of Cincinnati,
    19. LeeH. HeydrickC. DentonL. (1991) Effective reliability engineering during product development , Proceedings R&M CAE Concurrent Engineering Workshop,
    20. LeeY.J. SongJ. (2012) Finite-element-based system reliability analysis of fatigue-induced sequential failures. , Reliab. Eng. Syst. Saf., Vol.108 ; pp.131-141
    21. LevinM.A. KalalT.T. (2003) Improving product reliability and implementation., John Wiley & Sons,
    22. LinY.K. (2002) Using minimal cuts to evaluate the system reliability of a stochastic-flow network with failures at nodes and arcs. , Reliab. Eng. Syst. Saf., Vol.75 (1) ; pp.41-46
    23. OnoufriouT. ForbesV.J. (2001) Developments in structural system reliability assessments of fixed steel offshore platforms. , Reliab. Eng. Syst. Saf., Vol.71 (2) ; pp.189-199
    24. SafieF.M. FullerR.P. (2012) NASA applications and lessons learned in reliability engineering , Proceedings of the IEEE Annual Reliability and Maintainability Symposium, ; pp.1-5
    25. SalehJ.H. MaraisK. (2001) Highlights from the early (and pre-) history of reliability engineering. , Reliab. Eng. Syst. Saf., Vol.91 (2) ; pp.249-256
    26. SankaraiahG. SarmaB.D. UmasankarC. YadavalliV.S. (2011) Designing, modelling and optimizing of an integrated reliability redundant system. , S. Afr. J. Ind. Eng., Vol.22 (2) ; pp.100-106
    27. ShamsuzzohaA.H. HeloP.T. (2012) Development of modular product architecture through information management. , VINE: The Journal of Information and Knowledge Management Systems, Vol.42 (2) ; pp.172-190
    28. WilsonA.G. Anderson-CookC.M. HuzurbazarA.V. (2011) A case study for quantifying system reliability and uncertainty. , Reliab. Eng. Syst. Saf., Vol.96 (9) ; pp.1076-1084
    29. ZouT. MahadevanS. MourelatosZ. MeernikP. (2002) Reliability analysis of automotive body-door subsystem. , Reliab. Eng. Syst. Saf., Vol.78 (3) ; pp.315-324
    오늘하루 팝업창 안보기 닫기