【Weekly Topic】What is the difference between faults and failures? (2011.07.26–08.01)

Fault: A condition in which a product is unable to perform its designated functions. Except in the case of preventive maintenance or other planned activities, or a lack of external resources. A fault is usually a state that occurs after a product fails, but it may also exist before the failure. (GB/T3187-94) A abnormal condition that causes a functional unit to lose or reduce its ability to perform specified functions. (ISO/IEC DIS2382-14-96) An event or condition in which a product or a part of a product cannot or will not be able to perform its intended function. Certain products such as electronic components and ammunition are considered to be defective. (GJB451-90, GJB/Z91-97) Direct causes of failure (i.e.: mismatches, improper adjustment, defects, etc.). (MIL-STD-721C-81) Status as a cause of failure. (JIS Z8115-81) The condition in which a product fails to meet any of the requirements specified in the technical specifications and/or design (conceptual design) documents. (GOST27.002-89) (1) A physical condition that prevents a device, component, or element from being used in the intended manner, such as a short circuit, an open circuit, or an intermittent connection ； (2) Performance degradation due to misalignment, poor adjustment, inability to achieve proper alignment, or component failure ； (3) Direct cause of failure (e.g., poor adjustment, inability to adjust accurately, defects, etc.). (MIL-STD-1309D-92) 1. The state in which the product is unable to perform its specified functions. Except in the case of preventive maintenance or other planned activities, or a lack of external resources resulting in an inability to do so (amended from IEC 50:92). 2. Unplanned events or defects within the product, which can lead to the failure of the product itself or other related equipment, either one or more times (IEC50:92). Note 1: A fault is usually a state that occurs after the product fails, but it may also exist before the failure. Note 2: The product may contain defective units, which under certain conditions indicate a defect in the product itself (Definition 2 above). When these conditions occur, defects within the components will cause the product to fail, resulting in errors. This error spread to other products, also causing them to malfunction. After a failure occurs, overall, the product is faulty or in a faulty state (definition 1 above). (ECSS-P-001A-96) (1) Unforeseen circumstances that prevent a functional unit from performing its required functions ； (2) Manifestations of errors in software. Once a fault occurs, it may lead to failure and errors. (ANSI/IEEE STD 729-83) a. The functional component cannot perform the required function. b. In software, it represents an error, which is an explanation of a mistake. If encountered, it may cause a failure. Synonymous with hidden errors, defects (bugs). Note: The explanation of \"error\" as a mistake refers to the human actions that result in the creation of defective software. For example, missing or misinterpreting user requirements in the software manual, or incorrect translations or omissions of requirements in the design specification document. (GB/T11457-95) Performance degradation caused by component failure, drift, improper adjustment, and calibration. (MIL-STD-2155(AS)-85) Failure: An event in which a product loses its ability to perform the specified functions. (GJB3385-98) Events such as the product’s inability to perform its specified functions upon termination. (GB/T 3187-94) Termination of the ability of a functional unit to perform its specified functions. (ISO/IEC DIS2382-14-96) An event in which a product or a component of a product fails to perform the predetermined functions, or a state in which it is unable to function. (MIL-STD-721C-81) An event in which any part of the product fails to operate in accordance with its performance specifications. Failure may occur when it exceeds the minimum value required by the specifications, that is, when it goes beyond the design limits or safety thresholds. (Defense Acquisition Terminology-98) The product loses its specified functions. (JIS Z8115-81) (1) The termination of a functional unit’s ability to perform its functions. (2) An event in which the system or a system component fails to perform its required functions within specified limits under certain constraints. Failure may occur when a malfunction is encountered. (ANSI/IEEE STD 729-83) a. Loss of the ability of a functional component to perform its functions. b. The system or a component of it loses the ability to perform the required functions within specified limits. The system may fail when a malfunction occurs. c. The program operations deviate from the program requirements. (GB/T11457-95) Events that disrupt the product’s ability to function. (GOST 27.002-89) An event in which a product loses the ability to perform its intended functions. Note 1: The product malfunctions after failing. Note 2: “Failure” is an event, different from “Fault”, which is a state. Note 3: This concept does not apply to products composed of software. (ECSS-P-001A-96, IEC50(191)) The state in which a product is unable to perform its specified functions; failure is the manifestation of a malfunction. (MIL-STD-1309C-83) Changes in the operating characteristics of products that result in a decrease in useful performance. (MIL-STD-1309D-92) An event in which, under specified conditions and within a specified range, the product fails to perform one or more of the required functions. (MIL-STD-2155(AS)-85)

Reply to 2# ray1513: It would be better if you could illustrate with an example from your own organization.

When a machine loses its ability to function properly, it is considered to be faulty. When a machine part loses its ability to function properly, it is considered failed. Machine failures and component failures are inseparable; machines often fail due to the failure of their components. Therefore, important components should be carefully inspected; when it is detected that a component has failed or is about to fail, measures should be taken promptly to prevent accidents.

Difference between failure and malfunction: Failure refers to a situation where parts, components, equipment, or units lose their intended functionality. Fault: refers to a repairable failure phenomenon. In the event of a leakage fault in the mechanical seal of a pump, inspection and analysis reveal that the friction pair is broken; it cannot be repaired, resulting in failure of the seal and thus a replacement is necessary ； There were scratches on the end face of the static ring; after grinding, it was assembled and put into use normally. This is an example of a scratch fault occurring in the mechanical seal

Failure and Malfunction 1 Failure: An event in which a product loses its ability to perform its specified functions. (GJB3385-98) Events such as the product’s inability to perform its specified functions upon termination. (GB/T 3187-94) Termination of the ability of a functional unit to perform its specified functions. (ISO/IEC DIS2382-14-96) An event in which a product or a component of a product fails to perform the predetermined functions, or a state in which it is unable to function. (MIL-STD-721C-81) An event in which any part of the product fails to operate in accordance with its performance specifications. Failure may occur when it exceeds the minimum value required by the specifications, that is, when it goes beyond the design limits or safety thresholds. (Defense Acquisition Terminology-98) The product loses its specified functions. (JIS Z8115-81) (1) The termination of a functional unit’s ability to perform its functions. (2) An event in which the system or a system component fails to perform its required functions within specified limits under certain constraints. Failure may occur when a malfunction is encountered. (ANSI/IEEE STD 729-83) Editor’s note: This definition is also accepted by the International Organization for Standardization ISO. a. Loss of the ability of functional components to perform their functions. b. The system or a component of it loses the ability to perform the required functions within specified limits. The system may fail when a malfunction occurs. c. The program operations deviate from the program requirements. (GB/T11457-95) Events that disrupt the product’s ability to function. (GOST 27.002-89) An event in which a product loses the ability to perform its intended functions. Note 1: The product malfunctions after failing. Note 2: “Failure” is an event, different from “Fault”, which is a state. Note 3: This concept does not apply to products composed of software. (ECSS-P-001A-96, IEC50(191)) The state in which a product is unable to perform its specified functions; failure is the manifestation of a malfunction. (MIL-STD-1309C-83) Changes in the operating characteristics of products that result in a decrease in useful performance. (MIL-STD-1309D-92) An event in which, under specified conditions and within a specified range, the product fails to perform one or more of the required functions. (MIL-STD-2155(AS)-85) 2 Fault: A condition in which a product is unable to perform its intended functions. Except in the case of preventive maintenance or other planned activities, or a lack of external resources. A fault is usually a state that occurs after a product fails, but it may also exist before the failure. (GB/T3187-94) A abnormal condition that causes a functional unit to lose or reduce its ability to perform specified functions. (ISO/IEC DIS2382-14-96) An event or condition in which a product or a part of a product cannot or will not be able to perform its intended function. Certain products such as electronic components and ammunition are considered to be defective. (GJB451-90, GJB/Z91-97) Direct causes of failure (i.e.: mismatches, improper adjustment, defects, etc.). (MIL-STD-721C-81) Status as a cause of failure. (JIS Z8115-81) The condition in which a product fails to meet any of the requirements specified in the technical specifications and/or design (conceptual design) documents. (GOST27.002-89) (1) A physical condition that prevents a device, component, or element from being used in the intended manner, such as a short circuit, an open circuit, or an intermittent connection ； (2) Performance degradation due to misalignment, poor adjustment, inability to achieve proper alignment, or component failure ； (3) Direct cause of failure (e.g., poor adjustment, inability to adjust accurately, defects, etc.). (MIL-STD-1309D-92) 1. The state in which the product is unable to perform its specified functions. Except in the case of preventive maintenance or other planned activities, or a lack of external resources resulting in an inability to do so (amended from IEC 50:92). 2. Unplanned events or defects within the product, which can lead to the failure of the product itself or other related equipment, either one or more times (IEC50:92). Note 1: A fault is usually a state that occurs after the product fails, but it may also exist before the failure. Note 2: The product may contain defective units, which under certain conditions indicate a defect in the product itself (Definition 2 above). When these conditions occur, defects within the components will cause the product to fail, resulting in errors. This error spread to other products, also causing them to malfunction. After a failure occurs, overall, the product is faulty or in a faulty state (definition 1 above). (ECSS-P-001A-96) (1) Unforeseen circumstances that prevent a functional unit from performing its required functions ； (2) Manifestations of errors in software. Once a fault occurs, it may lead to failure and errors. (ANSI/IEEE STD 729-83) a. The functional component cannot perform the required function. b. In software, it represents an error, which is an explanation of a mistake. If encountered, it may cause a failure. Synonymous with hidden errors, defects (bugs). Note: The explanation of \"error\" as a mistake refers to the human actions that result in the creation of defective software. For example, missing or misinterpreting user requirements in the software manual, or incorrect translations or omissions of requirements in the design specification document. (GB/T11457-95) Performance degradation caused by component failure, drift, improper adjustment, and calibration. (MIL-STD-2155(AS)-85) Editor’s note: In GJB451-90 and GJB/Z91-97, no distinction is made between failures and faults. 3 Systematic failure: A malfunction caused by an inherent factor and occurring in a specific manner. It can only be eliminated by modifying the design, manufacturing process, operating procedures, or other related factors. Note: Repair work that involves no corrective measures generally cannot eliminate the cause of the fault. Systemic failures can be induced by simulating the causes of failures. (GJB451-90) Editor’s note: In GJB451-90, this definition is referred to as “systemic failure”. It must be related to some cause; it is a failure that can only be eliminated by modifying the design or manufacturing processes, operating procedures, documents, or other related factors. Repairive maintenance without modification measures usually cannot eliminate such causes of failure. This type of failure can be induced by simulating the cause of the failure. (GB/T3187-94) 4 Software failure: The inability to perform the intended logical operations in a specified/data environment due to defects in the software. (Defense Acquisition Terminology-98) 5 Single point failure: A localized failure that causes a product malfunction, with no redundancy or alternative procedures available as a remedy. (GJB451-90) Editor’s note: In GJB451-90, this definition is referred to as “single point of failure”. A single point of failure is a single hardware failure or software error that causes the product’s ability to perform its tasks to irreversibly drop below the level specified in the contract (the manner in which a product experiences a single point of failure is known as its single point of failure mode). (MIL-STD-1543B-88) The failure of a certain product will result in the failure of the system, and this cannot be compensated for by spares or alternative operating procedures. (MIL-STD-721C-81) 6 Catastrophic failure: A failure that results in casualties or the destruction of the system. (GJB451-90) Editor’s note: In GJB451-90, this definition is referred to as “catastrophic failure”. A sudden and complete failure mode that results in the loss of one or more basic functions. (MIL-STD-1309D-92) Failures that cause product loss. (MIL-STD-721C-81) 7 Fatal failure: A critical failure is a fault or combination of faults that prevents the product from performing its designated functions, or that could result in significant damage to people or property. (GJB451-90, GJB/Z91-97) Editor’s note: In GJB451-90, this definition is referred to as a “fatal fault”. A failure that may result in casualties, damage to important items, or other intolerable consequences. (GB/T3187-94) Failures or combinations of failures that prevent the product from performing its specified functions. (MIL-STD-721C-81) A failure or combination of failures that prevents the product from performing its intended functions or puts human lives at risk. (MIL-STD-1309D-92) 8 Random failure: A malfunction of a product caused by random factors. It can only be predicted through probabilistic or statistical methods. (GJB451-90) Editor’s note: In GJB451-90, this definition is referred to as “random failure”. Failures that can only be predicted through probability and statistical methods apply to all distributions. (MIL-STD-721C-81) It is not possible to predict its occurrence in an absolute sense; rather, any failure that may occur can only be predicted using probabilistic or statistical concepts. (MIL-STD-1309D-92) Product failure caused by accidental factors. (GJB/Z91-97) 9 Transient failure: A temporary failure caused by instantaneous or temporary external factors (such as fluctuations in the input power supply, rapid changes in ambient temperature, effects of electromagnetic waves, etc.) or by certain internal factors within the system. (GJB3385-98, MIL-STD-1309D-92) 10 Gradual failure: A failure that can be predicted through prior testing or monitoring; it results from the gradual change in the specified performance of a product as the number of service units increases. For electronic products, it is also referred to as drift failure. (GJB451-90) Editor’s note: In GJB451-90, this definition is referred to as “gradual failure”. Failure resulting from the gradual change in one or several parameter values of the product. (GOST27.002-89) Failure resulting from the gradual change over time of the performance characteristics specified for the product. This type of failure can be predicted through prior detection or monitoring, and sometimes it can be avoided through preventive maintenance. (GB/T3187-94) 11 Intermittent failure: A failure in which, after the product malfunctions, it resumes its normal function on its own within a limited time period, without the need for repair. (GJB451-90) Editor’s note: In GJB451-90, this definition is referred to as “intermittent fault”. Failure over a limited period of time, after which it can regain its ability to perform its specified functions without any remedial actions. (MIL-STD-721C-81) Faults of the same nature that occur repeatedly and can resolve themselves. (GOST27.002-89) 12 Soft failure: A failure in which performance is outside the specified range, but not to the extent that it results in a complete loss of functionality. (GJB3385-98, MIL-STD-1309D-92) 13 Common cause failure: Failure of multiple products resulting from the same cause. (Rewritten based on Reference 24) The failure of multiple products is caused by the same reason. (ECSS-Q-40A-96) 14 Common mode failure: Failure of multiple products caused by the same failure mode. Note: Common-mode failure is a special case of common-cause failure. (Rewritten based on Reference 24) The failures of multiple identical products stem from the same failure mode. Note: Common-mode failure is a special case of common-cause failure. (ECSS-Q-40A-96) 15 Potential failure refers to an identifiable condition that indicates the product or project will be unable to perform its specified functions. (GJB1378-92) Editor’s note: In GJB1378-92, this definition is referred to as “potential fault”. 16 Hidden function failure: A functional defect that cannot be detected by those using the equipment normally. a. For products that operate under normal conditions, any disruption in their functionality is not noticeable to those using the equipment. b. For products that operate normally, it is not obvious to those using the equipment whether they are functioning well when in use. (GJB1378-92) Editor’s note: In GJB1378-92, this definition is referred to as “hidden functional failure”. 17 Repeated failure pattern failure: Repeated failure refers to the occurrence of two or more failures of the same part under the same or equivalent conditions of use, with the underlying mechanisms behind these failures being identical; however, the incidence of such failures does not correspond to the expected failure rate for that part. (MIL-HDBK-781-87) 18 Multiple failure: Multiple failure refers to the simultaneous occurrence of two or more independent failures. If two or more components are found to be failed during fault detection, but it cannot be confirmed that there is a correlation between these failures, they are temporarily considered multiple failures. (MIL-HDBK-781-87) A fault event composed of two or more independent faults occurring sequentially, which may result in consequences that none of the individual faults could cause on their own. (GJB1378-92) Editor’s note: In GJB1378-92, this definition is referred to as “multiple failures”. 19 Independent failure/primary failure: A failure that is not caused by the malfunction of another product. (GJB451-90) Editor’s note: In GJB451-90, this definition is referred to as “independent fault”. A product failure that is not directly or indirectly caused by the failure or malfunction of another product. (GB/T3187-94) Failure that is not caused by the failure of any other product. That is, it is not a dependent failure. (MIL-STD-721C-81) is a failure that is not caused by another failure. (GOST27.002-89) 20 Dependent failure/secondary failure: A failure that occurs as a result of a fault in another product. (GJB451-90) Editor’s note: In GJB451-90, this definition is referred to as “dependent fault”. Failures caused by the failure of other related products. That is, it is not a independent failure. (MIL-STD-721C-81) Dependent failure refers to a situation in which an event causes two (or more) product items to be unable to perform their functions, which may render redundant measures ineffective and result in a single-point failure mode (SPFM); for example, damage to a cable conduit containing redundant power cables, or the impact of explosives causing several parallel relays to vibrate simultaneously. (MIL-STD-1543B-88) Failure of a product that is directly or indirectly caused by the failure or malfunction of another product. (GB/T3187-94) Failures caused by other failures. (GOST27.002-89) 21 Non-relevant failure: A failure that has been confirmed to be caused by use under conditions that are not in accordance with the specified requirements. Or it has been confirmed to be a fault caused solely by a certain design that will not be adopted. (GJB451-90) Editor’s note: In GJB451-90, this definition is referred to as “unrelated fault”. (a) it is determined that the failure is not caused by the specified operating environmental conditions, or (b) it is determined that the failure is inherent to the product design, which was not intended to be incorporated into the equipment in use. (MIL-STD-721C-81) Failures that should be excluded when interpreting test or operational results or calculating reliability metrics. The criteria for exclusion should be specified. (GB/T3187-94) 22 Relevant failure: A failure that must be taken into account when interpreting test or operational results, or when calculating reliability metrics. The criteria for inclusion should be specified. (GB/T3187-94) 23 Non-chargeable failure: A fault that is not related to any specific organization, or a fault that is explicitly designated as falling outside the responsibility scope of a particular organization. An example of distinguishing between responsible and non-responsible faults in practical work is shown in Figure 3. (GJB451-90) Editor’s note: In GJB451-90, this definition is referred to as “non-responsible fault”. (a) a non-related failure, or (b) a related failure that has been predetermined not to fall within the responsibility of that organization. All associated failures should be the responsibility of some organizational entity. (MIL-STD-721C-81) Figure 3 Example of classification of responsible and unresponsible failures 24 Responsible failure: A responsible failure refers to the associated or independent failures of the device under test, as well as any dependent failures resulting from them. These secondary failures are classified into a category and used to determine whether the acceptance or rejection criteria specified in the contract are met. (MIL-HDBK-781-87) 25 Primary fault: A fault that occurs in a unit when the stress on it is within the limits specified by the design standards, and is caused by factors inherent to that unit. (GJB768.1-89) 26 Dependent fault/Induced fault: A fault that occurs in a unit due to stresses exceeding the limits specified in the design. (GJB768.1-89) 27 Detectable fault: A fault that can be identified through testing, and for which an alarm or abnormal indication can be generated. (GJB3385-98) 28 Undetectable fault: A fault that cannot be detected using any testing sequence, due to circuit redundancy, the absence of testing points, or other reasons. (GJB3385-98) Faults that cannot be detected by any testing sequence due to circuit redundancy or the absence of testing paths. (MIL-STD-1309D-92) Editor’s note: MIL-STD-1309D-92 uses “impossible detects”. 29 Transient fault: an fault that becomes apparent only under the influence of certain environmental stresses. (GJB1032-90) 30 Soft fault: A fault that causes a reduction in the performance of the UUT ； Or a state that only appears under certain operating conditions of the UUT, and the fault disappears when these conditions change. (GJB3385-98, MIL-STD-1309D-92) 31 Delay fault: A fault in digital devices in which a level transition cannot occur within the specified time interval. (GJB3385-98, MIL-STD-1309D-92) 32 Latent fault: A fault that actually exists but has not yet manifested itself. (GJB3385-98) 33 Parametric fault: A fault that causes certain parameters of the equipment to exceed their specified range values. (MIL-STD-1309D-92) 34 Common cause fault: Failures of different components or different systems that are caused by a common underlying fault. (GJB768A FTA Guide Explanation) 35 Topologically independent faults: If the signal on a signal line associated with a certain fault affects the circuit, and when this set of signal lines does not intersect another set of signal lines related to another fault, meaning that these two faults do not affect the common parts of the circuit, then these two faults are considered topologically independent faults. (GJB3385-98) If it is determined that the number of faulty signal lines affects the circuit, and if one set of signal lines in that circuit does not intersect with another set, then these two faults are considered topologically independent; in other words, such faults do not affect the common parts of the circuit. As a result, topologically independent faults can be simulated simultaneously to accelerate the process. (MIL-STD-1309D-92) 36 Equivalent fault: Two or more faults that produce the same response to all possible tests. (GJB3385-98, MIL-STD-1309D-92) 37 Design fault: A fault resulting from improper product design. (GB/T3187-94) Failures that occur due to imperfect design rules and/or specifications for the scheme and structure, or as a result of violations of those provisions. (GOST27.002-89) 38 Manufacturing fault: A fault that occurs as a result of the product being manufactured in accordance with procedures that are not in line with the design or specified manufacturing processes. (GB/T3187-94) Failures that occur due to imperfect manufacturing processes or inadequate repair work carried out by repair companies, or as a result of violations of the relevant regulations. (GOST27.002-89) 39 Wear out fault/aging fault: A fault resulting from failures whose occurrence probability increases over time. It is the result of the product’s inherent processes. (GB/T3187-94, IEC50(191)-90) Failures that occur due to natural aging, wear, corrosion, and fatigue, under compliance with all applicable design, manufacturing, and usage rules and/or specifications. (GOST27.002-89) 40 Defect: A condition in which a product fails to meet the expected requirements for use or reasonable expectations. (GJB3385-98) Fails to meet certain expected usage requirements or reasonable expectations, including those related to safety. Note: The expectation must be reasonable under the existing conditions. (GB/T6583-94) Product defects, such as situations like: (1) omissions and imperfections detected in the early stages of the product’s life cycle ； (2) Defects contained in software that is fully mature enough to be tested or run. (Quoted from the IEEE standard “Guide to the Use of the Dictionary of Metrics for Developing Reliable Software”) 41 Error: The difference between a calculated, observed, or measured value or condition and the true, specified, or theoretically exact value or condition. Faulty products can cause errors. For example: a faulty computing device can produce calculation errors. (GB/T3187-94, IEC50(191)-90) Human activities that cause faults in software. For example: residual or misinterpreted user requirements in software specifications ； Incorrect conversion or omission of software requirements in the design specifications. (ANSI/IEEE STD 729-83) 42 Mistake: An act by a person that results in an undesired outcome. (GB/T3187-94)

Taking the thermocouple interlock as an example, an interlock failure may be caused by a loose connection in the thermocouple wire, leading to incorrect readings； However, a failure in the interlock can be caused by a broken thermocouple wire or desulfurization, which results in no display or an incorrect measurement range, thereby triggering the interlock action.

Failure refers to the situation in engineering where a component loses the function specified by its original design. Failure includes a complete loss of the original function ； Functionality is reduced, and there are severe damages or potential risks; continuing to use it will result in a loss of reliability and safety. Fault: A phenomenon in which a device, during operation, \"loses its specified functions\" or endangers safety for some reason. Specified functions refer to the functions clearly defined in the technical documentation of the equipment. Failure is sometimes also referred to as a malfunction; losing data while a device is in use can also be considered a malfunction, but these issues can be repaired.

Malfunction and failure are two completely distinct concepts, and the consequences they bring about are also vastly different. For example: If the UPS power supply of the DCS system fails, the system activates its backup batteries (for 10–30 minutes), and an alarm is triggered. Maintenance personnel can use the prompts to locate the problems and resolve them, thereby maintaining continuous production. If it fails, that would be terrible; although a backup battery is available to keep it running, the alarm stops working and no one knows until the system shuts down. Or if the backup battery fails, it will directly lead to a cascading shutdown of the system. There are countless examples. Only by conducting regular inspections and periodic tests to eliminate potential hazards can we ensure that there is a possibility of malfunctions, but no chance of complete failure............ This is just my humble opinion; it’s for reference only.

Start-up shut-off valve malfunction: Due to electrical reasons, the normal pneumatic valves within the unit enter a malfunction state; the start-up shut-off valve at the pump inlet closes completely, while the purge gas shut-off valve opens fully. The heater control valve was fully closed, causing partial shutdown of the plant! It took nearly two hours of hard work to get things back to normal

Fault: The condition in which a product is unable to perform its specified functions. Except in the case of preventive maintenance or other planned activities, or a lack of external resources. A fault is usually a state that occurs after a product fails, but it may also exist before the failure. Failure: An event in which a product loses its ability to perform its specified functions. Therefore, my conclusion is that a failure can lead to a malfunction, but a malfunction isn’t necessarily caused by a failure! Prompt repair of non-fatal faults can prevent significant losses, while failures that lead to system breakdowns can result in substantial losses!