Reliability Engineering Literature Review and Research Questions - Assignment Example

Question 1: The concept quality and requirement has been antagonistic and that has also been expressed in Rethinking the Notion of Non-Functional Requirements (Glinz 2005). The issue which has been brought out by the author is the relationship between quality and requirement and how these integrate themselves with notion of non-functional requirements. However, there are issues raised that need further comment. Firstly, the author admits that there are two classes of requirements; functional and non-functional requirements. Reacting to non-functional requirements, he argues that this is something to do with quality, somehow soft, mostly qualitative. This is the point of my disagreement with the article. To express my concern, his way of classifying the two types of requirements clearly shows separating concerns. That is, it avoids the issue that a requirement can be non-functional; more so when stated in qualitative form. To me, this already shows that requirement and quality are not only working together but also not distinct. A good example of my argument is that there is a possibility of one requirement to be functional if they are stated operationally and the same requirement becoming non-functional if expressed qualitatively. Further to this, the author represents facet of the classification which can help in identifying the proper type of verification for a given requirement. This representation to my view does not augur well with his definition of quality as cited from ISO 9000:2000 (ISO 9000, 2000). For instance, on the first hand, he presents that an operational requirement can be verified through acceptance testing. On the other hand, he works with the definition that, “…a requirement is a need or expectation that is stated, generally implied or obligatory” (Glinz, 2005, p.1). Integrating his understanding of requirement and quality, it can be argued that the two concepts are closely intertwined. This is the point I agree with the author. Sommerville (2004) also shows that requirement and quality are closely intertwined but the relationship is what he refers to behavioural vs. non-behavioural requirements. Another interesting point that needs scrutiny is the argument that there is a possibility of classifying requirements according to their kinds. To my understanding, there needs to be discrete understanding that requirement may concern a function or data---as a matter of fact this is what Glinz (2005) explains as functional requirement. On contrary, he further argues that the same requirement can represent or specify a given quality like reliability. While the author tries to show that the two concepts are distinct, this is not true owing to the fact that requirement can be viewed in terms of quality as well (Wieringa 2000). Giving practical example, assuming a company wants to establish its strategies therefore the machine or the system the company uses will execute order tracking process to enable the company prioritises on the strategies. Question 2: Significance of importance measures is multifaceted. While Barabady and Kumar (2006) discuss this issue, researches highlighting such significance are equally dynamic. To understand this issue from its totality, it is necessary defining importance measures within the context of a component. According to Owens et al. (2006), importance measures is a partial derivative of a given system availability with regard to its failure rate, availability, and repair rate. Most articles reviewed by Barabady and Kumar thus far base significance of importance measures on the elements of its definition---repair rate, availability and failure rate. To begin with is what Barabady and Kumar (2006) explain to be the system reliability and availability. This is actually embedded on the available researches. These researches, as cited by Barabady and Kumar, argue that reliability and availability are good evaluations of a system’s performance which is actually brought about by of importance measures (Owens et al., 2006; Elegbede and Adjallah 2003). How does the argument become practical in real life situation? This does so where there is a system undergoing design stages that will help determine or estimate the needed reliability and availability of each of its components. It is common issue among systems that their values and output of their structures are dependent on their ages---which also affect their reliability and availability. Therefore, the requirements for these systems to be operational are reliability and or availability or what can be termed as mean time between the system’s failures (BTBF). This is actually where importance measures come in. BTBF is not only the thing here but also mean time to repair the same machine. This argument can be justified by bringing the aspect of criticality of maintainability and reliability features of systems especially those at component level. In this case or in these systems, it will be prudent to consider reliability and maintainability importance measures since this will ultimately improve availability characteristics of the given system. How is this example justified in researches? Theoretical problems of availability optimisation and allocation using different methods have been presented by Chiang and Chen (2006) and Elegbede and Adjallah (2003). Another significance of importance measures as presented by Barabady and Kumar (2006). They argue that the process ensures that weakest areas of a machine or system are identified. Further to this, it also ensures that there are modifications to these systems which ultimately improve maintainability and reliability. There exist a number of systems developed in this line to test importance measures in the reliability area. Practical example that can be justified is the crushing plant in Jajarm Bauxite mine of Iran. Actually, researches from scholars such Chiang and Chen (2006) have brought the introduction of mathematical concept of the importance measures which have been of great significance to Jajarm Bauxite mine. Assessment of plants such as Jajarm Bauxite mine in terms of its performance entirely depends on its components. It can be noted that this plant has some sub-systems which practically influences its availability and reliability than other sub-systems within it. Therefore, for the company or engineers to evaluate how some sub-systems are more reliable and available than others, importance measures become handy. Barabady and Kumar (2006) add that as far as understanding these sub-systems are concerned, importance measures also necessitate Reliability Criticality Importance, Birnbaum Component Importance and Structure Importance. Actually, Birnbaum’s component importance is one of the commonly used reliability importance indices (Frickes and Trivedi, 2003). How is this done practically? Take for instance Jajarm Bauxite; when its failure characteristics are ascertained, its reliability importance can be estimated based on such characteristics. Actually, mathematical expressions have been derived to help arrive at calculations involving failure characteristics of machines or systems (Elegbede and Adjallah 2003). Availability importance measure based on the repair rate/MTTR and Availability importance measure based on the failure rate/MTBF are other significances of importance measures. Beginning with the MTTR, importance measures help engineers ascertain repair rate of a plant on its availability. On the other hand, importance measures also help understand failure rates in the availability of the whole system. Let us assume the case of Electricity Transmission Systems (ETS) which have components interconnected to transport electricity energy from the bulk transmission systems to various load points. In these systems, MTTR and MTTR are therefore used to identify weak links so as to avoid repair and failure rates---proposed measures concern the outage rate of the system and its sub-components rather than the probability of failure or survival. Question 3: Emergent properties bring one of the core argument and significance challenges affecting engineers of complex systems. Generally, this concept is looked into in terms of unexpected behaviours that interact between the components of an application and their environment. Linking the concept to availability, reliability and maintainability, it needs to be understood, from the view point of literatures reviewed that emergent properties can either be beneficial or harmful (Owens et al., 2006). Basing on its benefits, engineers or users can adapt to products to support tasks that designers never intended. On the other hand, it can be harmful in the sense that they are likely to undermine vital system requirements. While this appears so, there is however, disagreements regarding the nature of emergent properties. To understand this disagreement, adopting Sommerville (2004) ways of looking at systems can help. To him, complex systems have different characteristics. Actually, he defines complex systems as, “the ones that contain many components and layers of subsystems with multiple non-linear interconnections that are difficult to recognise, manage and predict” (p.34). His definition can therefore be linked to emergent properties by arguing that the concept can be used to discern or distinguish complex systems from application that are merely complicated. Actually here is where the concept “operational feasibility of a system is an emergent property of a system” is best suited. Contemporary authors such as Chiang and Chen (2006) have approached the concept by first attacking complex systems. A good example they raise is that unexpected patterns of behaviour do not reveal special forms of ‘emergent’ properties. Rather, what is done is just illustration of problems with our understanding of complex systems. Chiang and Chen’s critics augur well with modern theories of emergence since these theories too seem to ask the same questions---the perspective of the person making the predictions. Prediction within this context mean designers and engineers fail to make exact knowledge regarding systems and their expected environment of operation. Therefore, the assumption made here is that emergent properties are those that cannot be predicted by individuals having knowledge regarding features of the parts of a complex system and its environment. Perhaps this is why Chiang and Chen brought the concept weaker and stronger emergence in understanding reliability, maintainability and availability. For instance, weak emergence engineers are able to get higher levels of behaviours of a system thus enhancing maintainability. They further link strong emergence with system reliability which they argue depends on components of that systems and thus unexpected interactions are likely to cause failures thus affecting reliability. It seems definite; therefore that emergence or emergence properties have significance for the design and engineering of many systems and applications. It is not clear though, that ideas around emergence can make vital contribution to design and engineering. Question 4: Reliability, availability and maintainability are related terms as far as systems are concerned. Beginning with reliability, this is a measure of probability that a machine will perform its given or designed function for a given specified period of time (Elegbede and Adjallah 2003). This is related with availability which is the relative measure of how simple it is to restore the system. Wieringa (2000) relates these two terms with maintainability which is the measure of how long equipment can take to be restored back to its availability. Definitions of these terms have been revisited to helped form basis of the argument regarding through-life aspects. Past researches have documented that systems can be unstable as a result of reliability, availability and maintainability (RAM) issues which are identified during development, operational and testing period especially when RAM is traded out. Connecting the implications of not focusing on the through-life aspects and system availability; availability is concerned with the readiness of any system or machine to be committed to task. Therefore failing to check how available system(s) derails its level of operation--- Wieringa (2000) adds that logistic delays are clear indicators that through-life is compromised when availability of a machine is not checked. As stated earlier, availability, maintainability and reliability are two related terms. While availability enables a system to start a mission, reliability makes it complete the mission and maintainability will enable restorability of availability. The translation of the above statement is that if sufficient resources are committed to task then all levels of RAM will be achieved even though Wieringa (2000) warns that the cost will be higher than the expected optimum level. Generally, the higher the system reliability, the lesser the logistic and maintainability of a system becomes and the lesser the cost of maintaining a system at any given availability level. This is when a machine or system can halt when there is failure to focus on the through-life aspects. Need to carry out RAM assessment of a system as it becomes more and more complex. For instance, there are reliability modeling techniques such as evaluation, control and predictions which necessitate proper design, effective maintenance and dependable operations of systems. In as much, engineers continue to face challenges of design redundancy. In most cases, small variations such as parameters in RAM of system redundancies can be highly sensitive and only through precise modeling of RAM and deep understanding of RAM reports that one can eliminate challenges of not focusing on the through-life aspects. There is increase in need of the reliability, availability and maintainability assessment among various organizations. Reasons documented for such actions are that they help in understanding and appreciating the potential impact of failures and downtime for the system. Regardless of the role of the system, it is reasonable to assume and state the degree of a service or product success as this is directly related to its ability to meet and surpass average customer expectation. Another implication of not focusing on the through-life aspects relates to reliability and manufacturing machines. Focusing on through-life aspects means carrying out reliability prediction. Reliability prediction on the other hand, helps necessitate the proposal of a design and which ultimately provide a quantitative basis of selection among various competing component or approaches (Owens et al., 2006). Predicted results can be used in ranking and assessing design problems of systems. During implementation of process of the system reliability, there is need for continuously testing for system from the beginning of its development. This will enable uncovering and correcting of reliability problems as they are realized. Therefore not focusing on the through-life aspect---reliability testing means problems will not be uncovered thus causing break down. To this regard, Frickes and Trivedi (2003) add that reliability program for equipment, system production and development are part of growth and development and best practices to be adopted by an organization. The success or failure of a system is directly related to the management and maintenance of RAM related issues; including prevention failures during its design phase. This should be a continuous process throughout long life process of a system. Lack of continuous long life RAM maintenance exposes system to failure and initial operational MTBF failure hence lack of system growth. Lack of continuous RAM maintenance and improvement during its design stages also result into a very low initial MTBF and low growth potential as these are the main reasons why a system may fail to meet its operational suitability. As Sommerville (2004) concludes in his research, RAM is very important to various suitable factors which include logistic support and training. Dependable failure prevention always starts with reliable growth in the design phase of a system. A system which is fielded with specific RAM shortcomings usually increases cost of maintenance as well as the repair cost. References Barabady, J. and Kumar, U. (2006). Availability allocation through importance measures. Division of Operation and Maintenance Engineering, Lulea° University of Technology, Lulea°. Chiang, C-H. and Chen, L-H. (2006), “Availability allocation and multi-objective optimization for parallel-series systems”, Reliability Engineering and System Safety, available at: www.scincedirect.com (in press). Elegbede, C. and Adjallah, K. (2003), “Availability allocation to repairable systems with genetic algorithms: a multi-objective formulation”, Reliability Engineering and System Safety, Vol. 82 No. 3, pp. 319-30. Frickes, M.F. and Trivedi, K.S. (2003), “Importance analysis with Markov chains”, Proceedings of the 49th Annual Reliability and Maintainability Symposium, Tampa, FL, 27-30 January, pp. 89-95. Glinz, M. (2005). Rethinking the Notion of Non-Functional Requirements. Department of Informatics, University of Zurich Winterthurerstrasse. ISO 9000 (2000). Quality Management Systems – Fundamentals and Vocabulary. International Organization for Standardization. Owens, J., Miller, S. and Deans, D. (2006), “Availability optimization using spares modeling and the six sigma process”, Annual Reliability and Maintainability Symposium, pp. 636-40. Sommerville, I. (2004). Software Engineering, Seventh Edition. Pearson Education. Wieringa, R.J. (2000). The Declarative Problem Frame: Designing Systems that Create and Use Norms. Proceedings of the Tenth International Workshop on Software Specification and Design. San Diego. 75-85. Read More