| | | Reality
 |
|
Arguably this type of bounded rationality perspective may in part explain why the noted physicist and academic Professor Arie Dubi (14) has come to form the view that "System Engineering was, for many years an art composed of an eclectic collection of methods and procedures. A vast array of excuses ranging from lack of data to excessive unnecessary complexity was used to justify semi heuristic solutions based on uncontrolled approximations." In his experience, exponential distributions are assumed in almost all cases of reliability and availability analysis, detaching the problem from ageing and maintenance issues. Where maintenance is required, immediate repair is assumed. In spare part models homogeneous Poisson processes are assumed and no relation is established between resources (such as spare parts) and system performance. "Most categories of system engineering are treated as independent subjects, such that Availability analysis is done as if no spare part problem exists. Maintenance analysis is executed with no relation to resources and performance and non-Markovian analysis is almost unheard of." |
|
Whilst this may seem like someone "taking a pop" at systems engineers, the view expressed by Arie Dubi is absolutely representative of my experience to date. |
|
Dubi's view is effectively supported in Engineering Reliability whereby Richard Barlow (15) questions whether the use of classical statistics is logically tenable. He contends that classical statistics is based on deductive analysis (the logic of mathematics), whereas statistical inference and decision theory are concerned with inductive analysis (probability judgments). He quotes Basu (16) who said, "It took me the greater part of the next two decades to realize that statistics deals with the mental process of induction and is therefore essentially antimathematics. How can there be a deductive theory of statistics?" Therefore the ability to forecast availability goes beyond statistical analysis of the reported times between failures, the corrective and preventive maintenance times and the logistic delay times. However it is dependent upon the empirical data provided by such reporting and therefore the two are inextricably linked. Forecasting examines empirical data, building a picture of individual equipment and fleet behaviours that can be scientifically analysed to mathematically express observed relationships between reliability, operational availability and support factors to calibrate the probability of future operational success and cost impact. |
|
So, why am I particularly concerned about Availability? Why not Reliability to the same extent? Well I am concerned about Reliability but it comes from the abstract world of design. In Reliability Simplified: Going Beyond Quality to Keep Customer for Life, Harrington et al (17) make the case that Intrinsic Reliability is defined by the design. I wholeheartedly agree. |
|
We know, particularly for complex items, that although an item is a function of system design, the item does not generally of itself constitute a system (18). Additionally, the probability that an item will perform its required functions for the duration of a specified mission profile is dependent upon its use up to the point that it embarks on the mission, and the availability and opportunity to apply the resources required to maintain it for the period between that point and the planned mission completion. Therefore in itself, the suitability and usefulness of reliability as a measure of effectiveness is limited. |
|
People also use the terms Intrinsic and Inherent Reliability, probably because if we consult most dictionaries the terms Inherent and Intrinsic are regarded as wholly interchangeable. I take a different view in this respect because I have always regarded that something Intrinsic has purity, the essence of something, rather than Inherent, which is closely connected with, stems from, or holds onto the Intrinsic nature but does not necessarily perform or react exactly as expected in terms of the Intrinsic value. The difference between the two terms is borne out in various etymological dictionaries. Therefore, I consider that insofar as Inherent Reliability is concerned, we have crossed over the line into the domain of Availability. |
|
Simply put I live in the real world, the physical world, where I have to deal with the physical truth, the practical consequences of Reliability, Availability, because I want to make an effective and efficient contribution to the outputs of my organization. |
|
Availability can be defined as the probability that the system is operating properly when it is requested for use (Instantaneous or Point Availability). That is, availability is the probability that a system is not failed or undergoing a repair action when it needs to be used. |
|
But the definition of availability seems to be flexible. As a result, there are a number of different classifications or types of availability; Inherent, Achieved, Instantaneous, Steady State and Operational as examples. The different classifications are largely based on the type of downtime one chooses to consider in the analysis. |
|
Blanchard (19) says that Availability is viewed as "the measure of the degree a system is in the operable state at the start of a mission when the mission is called for at an unknown random point in time. This is often called operational readiness." In fact it can be termed in many different ways. For example Airbus use the term Operational Reliability (20) to describe what in essence is Availability." Like James V Jones (21), I am of the view that "The mere fact of possessing a system provides no value to the owner. The value of possessing a system comes through the ability to actually use it. Therefore the ability to use a system becomes the final gauge of its value to the owner. The most common term for this value of a system is availability, or the ability to use a system when required." He makes the point that conceptually it is not enough to own a car; it is the ability to have it provide transportation upon demand. If the car is inoperative, unavailable, because of some failure, or has no fuel or tyres, or driver for that matter, then it cannot provide its necessary value contribution to transportation capability. Therefore it can be argued that the concept of availability has been developed as a gauge of the value and a measure of the effectiveness of a system. Availability is a metric; it is a performance criterion for repairable systems that accounts for both the reliability and maintainability properties of a component or system. |
|
What is consistent and clear from the multiplicity of classifications and definitions, other than Point Availability, is that Availability is concerned with probability. From the users' perspective perhaps the most important of the classifications and types is Operational Availability because at this stage we are delivering our Capability for a purpose, as a means to deliver outputs. |
|
last update: January 10, 2006 |
|
|
| | |
