There is no generally accepted definition of the reliability of a technical product. The definition and interpretation of the term vary from industry to industry and from user to user. For the purpose of this book, we choose a rather wide definition of the reliability of a technical item.
The ability of an item to perform as required in a stated operating context and for a stated period of time.
The term item is used to designate any technical system, subsystem, or component. The items studied in this book are built of hardware parts, and to an increasing degree, of software. When relevant, the user interface is part of the item, but operators and other humans are not part of the items studied here.
The reliability concept is illustrated in Figure 1.1. The required performance is determined by laws and regulations, standards, customer requirements and expectations, and supplier requirements, and is usually stated in a specification document, where delimitations of the operating context are stated. As long as the predicted performance at least fulfills the required performance, the item is reliable – when it is used in the same operating context and for the period of time stated in the required performance.
Figure 1.1 The reliability concept.
By operating context, we mean the environmental conditions the item is used in, the usage patterns, and the loads it is subjected to, and how the item is serviced and maintained.
Definition 1.1 is not new and is not created by us. Several authors and organizations have used this, or a very similar definition of reliability, at least since the 1980s. A more thorough discussion of reliability and related concepts is given in Section 1.3.
1.1.1 Service Reliability
A service is provided by a person, an organization, or a technical item to a person or a technical item. The entity providing the service is called a service provider, and the entity receiving the service is called a customer. Services can be provided on a (i) continuous basis (e.g. electric power, computer networks), (ii) according to a timetable (e.g. bus, rail, and air transport), or (iii) on demand (e.g. payment by debit cards).
Many services are provided by a single service provider to a high number of customers. A customer considers the service to be reliable when she receives the service (e.g. electric power) with sufficient quality without outages. We define service reliability as follows:
Definition 1.2 (Service reliability)
The ability of the service to meet its supply function with the required quality under stated conditions for a specified period of time.
Several quantitative service reliability metrics have been defined, but they vary between the different types of services.
1.1.2 Past and Future Reliability
In our daily language, the term “reliability” is used to describe both past and future behavior. We may, for example, say that (i) “my previous car was very reliable” and (ii) “I believe that my new car will be very reliable.” These two statements are quite different. The first statement is based on experience with the car over a certain period, whereas the second statement is a prediction of what will happen in the future. We distinguish them by using two different terms.
Reliability (single word) is always used to describe the future performance of an item. Because we cannot predict the future with certainty, we need to use probabilistic statements when assessing the reliability.
Achieved reliability is used to describe the item's past performance, which is assumed to be known to the analyst. No probabilistic statements are therefore involved. The achieved reliability is also called observed reliability.
The focus of this book is on reliability and the future performance. The achieved reliability is most relevant in Chapter 14, where analysis of observed failure data is discussed.
1.2 The Importance of Reliability
Several producers of technical items have struggled and even collapsed because of item flaws and failures. To build a reputation for reliability is a long‐term project, but it may take a short time to lose this reputation. The main drivers for high reliability are listed in Figure 1.2. Over the years, the reliability has improved for almost all types of items, but at the same time, customers expect a higher and higher reliability of the new items they buy. Current customers further expect that possible failures in the warranty period are rectified without any cost to the customer. To be attractive in the market, the suppliers have to offer a longer and longer warranty period.
Figure 1.2 Main drivers for high reliability.
If items have flaws that affect safety, safety regulations may require all the flawed items to be recalled for repair or modification. Such recalls are rather frequent in the car industry, but are also common in many other industries. In addition to excessive warranty costs and item recalls, flawed items lead to dissatisfied and nonreturning customers.
1.2.1 Related Applications
Reliability considerations and reliability studies are important inputs to a number of related applications. Several of these applications have adopted the basic terminology from reliability. Among the relevant applications are:
Risk analysis. The main steps of a quantitative risk analysis (QRA) are: (i) identification and description of potential initiating events that may lead to unwanted consequences, (ii) identification of the main causes of each initiating event and quantification of the frequency of the initiating events, and (iii) identification of the potential consequences of the initiating events and quantification of the probabilities of each consequence. The three steps are shown in the bow‐tie model in Figure 1.3, where the main methods are indicated. The methods that are covered in this book are marked with an .
Maintenance planning. Maintenance and reliability are closely interlinked. High‐quality maintenance improves the operational reliability and high reliability gives few failures and low maintenance cost. The close link is also visible in the popular approach reliability‐centered maintenance (RCM), which is discussed in Chapter 9.
Quality. Quality management is increasingly focused, stimulated by the ISO 9000 series of