ABSTRACT
Products should be safe to use This means that any product should be designed such that it does not cause accidents and generate any adverse health effects (eg, exposure to harmful conditions or toxic substances) on people during the product life cycle Accidents can cause injuries, loss of lives, loss of work, delays in work, property damage, and also can incur costs Designers should ensure that a safe product is created in the early conceptual and detailed design stages so that failures of the product during its uses are minimized A safe product should have characteristics and features that prevent accidents from occurring and prevent injuries in cases if accidents occur Just like the proactive approach to quality, safety should be “builtin” the product in its early design stages and safety evaluations (or monitoring) must be conducted as the product advances through its life stages
Customers are very conscious of safety and they demand that products should be safe to use, and the users and others should be free from any harm that can result from the uses and even some foreseeable misuses of the product Thus, safety is an important product attribute and it should be considered at the earliest time in the conceptualization of any new product Product liability is also another important consideration during the product design Safety requirements should be developed early to ensure that safe products can be developed and potential for the product liability related to costly litigations are minimized Special surveys of new safety technologies, benchmarking of safety features in other similar competitive products, safety reviews, and safety analyses should be conducted during the early design stages
The objective of this chapter is to provide basic background into Safety Engineering, related issues and considerations, methodologies, and safety-related costs to ensure that safe products are developed The topics covered in this chapter include the following: (1) definition of accident, (2) accident causation theories, (3) safety performance measurement, (4) product liability principles, and (5) approaches and methodologies used in solving safety-related problems
Safety Engineering is a specialized engineering field that deals with application of multidisciplinary concepts and techniques to design and evaluate products, systems,
and processes with the primary objective of improving safety and providing healthful working environments
People build work systems (where different types of work gets done-from producing products to providing services) involving workstations that include equipment (products), people (workers, operators, or users), and environment Some workplaces may have unsafe working conditions and some people may commit unsafe acts The products could also be used by people in unsafe conditions and/or while committing unsafe acts The unsafe acts (committed by people) and unsafe conditions can cause accidents; and the accidents can cause injuries and losses (incur costs) The accidents can also lead to work stoppages and inefficiencies in work processes Many research studies involving analyses of the accident data and accident causation factors have shown that the majority of accidents are preventable (Heinrich et al, 1980) Product manufacturers incur safety costs due to (1) litigations and liabilities resulting from the accidents caused by their products, (2) extra efforts undertaken to create safer products through accident prevention actions (eg, incorporation of safety features in the products and conducting special design reviews and tests during product development to ensure that the products are safe), and (3) accidents during the life cycle of the products The safety-related costs of a product manufacturer can amount to about 5% to 15% of the revenues generated by the products (eg, add all the costs in conducting literature surveys, benchmarking, safety analyses, safety reviews, purchasing, testing, installing safety devices, defending litigations, etc)
The need for safety engineering generally becomes obvious when any of the following problems occur: (1) increase in number of accidents and injuries, (2) increase in the costs or losses resulting from the accidents, (3) employee turnover, (4) employee complaints, and/or (5) increased product litigations
Figure 111 illustrates the safety engineering approach in dealing with existing or new products The safety problem is generally noticed with the mounting accident statistics obtained from the accident databases, changes in the safety regulations, increases in customer complaints, and/or product liability litigations The data from these sources are reviewed by the product designers and engineers to determine how various safety problems can occur Many accident causation theories are considered and accident situations are analyzed to determine the causes of the accidents Several hazard analysis techniques (see Chapter 16) are also applied to predict safety critical situations before any accidents occur The analyses uncover many potential causes for future accidents that could be eliminated by undertaking one or more accident preventive countermeasures Costs/benefits analyses are usually performed to determine the countermeasures that can provide higher ratios of the benefits (ie, reduction in costs due to preventable accidents) to the costs of implementation of the countermeasures The selected countermeasures are implemented and the safety performance is monitored (eg, by maintaining control charts of the accident data) to determine the effectiveness of the countermeasures If the countermeasures are found to be ineffective, then further changes in safety prevention strategies are considered by iterating the whole process as shown by the feedback loop in Figure 111
Safety engineering is an important field as it seeks to: (1) design safer products and (2) build and operate safer systems Safer products and systems are generally efficient to use as they tend to reduce costs and time losses due to accidents Thus, safety engineering should be considered as one of the important disciplines during the entire product life cycle, and it is especially useful during the early phases of product and system development
Goetsch (2007) has pointed out that need for safety engineering becomes more acute when any one or more of the above problems are observed in an organization: (1) rapidly increasing safety-related costs, (2) need to meet new safety regulations, (3) increasing litigations, (4) growing interest in ethics and corporate responsibility, (5) increased pressure from labor organizations and employees, (6) realization that safer products and systems are more productive, (7) realizing that safety and quality are closely related, (8) greater awareness and professionalization of health and safety, and (9) new hazards due to faster pace of technological changes
In developing safe products, engineers should always think about the “3Es” of safety engineering, which include the following:
1 First “E” for Engineering: Products should be designed and engineered with safety in mind
2 Second “E” for Education: Safety education will help the designers and users to understand the importance of safety
3 Third “E” for Enforcement: Safety requirements and safety practices must be enforced (through approaches such as training, incentives, inspections, audits, regulations, fines or penalties, and product recalls) to ensure that people act responsibly
Thus, commonly considered safety countermeasures include the following:
1 Engineering solutions (eg, incorporation of fail-safe designs, lock-outs, and alterations to products and processes to minimize accidents and injuries)
2 Administrative solutions (eg, screening employees, limiting exposures to unsafe/toxic environments, rotating people, establishing policies, regulations/laws, practices/procedures, enforcement, training, and awareness)
3 Personal protection (eg, isolation of people from hazards, providing hard hats, safety goggles, and masks, seat belts)
The techniques of safety engineering should be applied by engineers in designing complex products and systems to minimize the probability of safety-critical failures The “Systems Safety Engineering” function helps to identify “safety hazards” in emerging designs and can assist with the techniques to “mitigate” the effects of potentially hazardous conditions that cannot be designed out of systems
The methods used by safety engineers are listed as follows:
1 Critical incident technique 2 Behavioral sampling 3 Checklists to identify hazards 4 Hazard analysis (or methods safety analysis) 5 Fish diagram (cause and effect diagram) 6 Failure modes and effects analysis (FMEA) and failure modes, effects, and
criticality analysis (FMECA) 7 Logical analyses and fault tree analysis (FTA) 8 Reliability analysis 9 Risk analysis 10 Cost/benefits analysis 11 Accident data analysis (eg, data gathering and statistical analysis of acci-
dent frequency, rates, severity, and accident costs) 12 Accident investigation 13 Accident reconstruction and accident simulation 14 Control charts (of different types of hazards, unsafe acts, unsafe conditions,
and accidents) 15 Experimental studies (eg, to determine effects of countermeasures on
near-accident and accident rates by comparing “before” the countermeasures to “after” the countermeasures data)
The first 10 of the methods listed above can be conducted without waiting for accidents to occur The accident data-based methods (#11, 12, 13, and above) can be applied after the accidents have occurred The control charts and experimental studies can be based on measurements of non-accidents (eg, unsafe acts, unsafe conditions, hazards, or near-accidents) or accidents The above methods are described in Chapters 13 and 16
Some historic points in the evolution of safety engineering are briefly described below
1 Circa 1800: Before the industrial revolution in Britain and the United States (circa 1800), claims of injury or damages were predicated on the laws of contracts/privities and trespass (Hammer, 1980) (The rule of privity can be described as follows: A seller is liable for injury by his product only to the party with whom he has contracted to supply the product)
2 1830-1910: Industrial Revolution (after circa 1830), Factory Inspections (1835-1875) (eg, guarding of dangerous machines), 1880 Boiler Standards and 1910 Workmen’s Compensation Acts played major roles in advancing safety in the industry (The workmen’s compensation can be described as: “No fault” liability which limits damages an employer has to pay for an injury at work and limits the right of employees to sue their employers)
3 1925-1960: Safety Research was primarily conducted by the insurance companies (responsible for settling injury-and accident-related claims covered by the insurance policies)
4 1963-1970: A number of government safety laws were enacted by Congress (eg, the Consumer Products Safety Act and the Motor Vehicle Safety Act) Many federal safety agencies (eg, the Consumer Products Safety Commission and the National Highway Traffic Safety Administration) were created to develop and enforce safety regulations
5 1963-1973: A number of product safety and liability cases and decisions (eg, strict liability) (Note: Product Liability concepts are presented in a later section entitled “Product Safety and Liability” of this chapter)
6 1973-present: Strong awareness of safety issues due to presence of federal and state safety regulations, enforcement activities, and well-maintained accident databases by government agencies and private organizations Thus, safer products are now expected by all
A number of safety researchers have provided definitions to describe an accident, that is, when a situation would be called an accident The definitions also help in understanding the concepts of an accident and issues related to accident prevention A few commonly referred accident definitions are provided below
1 An accident is a set of complex events involving sequence, human actions/ behavior (unsafe acts), unsafe conditions, and some degrees of the following characteristics (Petersen and Goodale, 1980):
a Degree of unexpectedness-the less an event could have been anticipated, the more it is likely to be called an accident
b Degree of avoidability-the less the event could have been avoided, the more likely it is to be called an accident
c Degree of intention-the less the event resulted from a deliberate action or lack of an action, the more likely it is to be called an accident
d Degree of warning-the less warning, the more likely it is to be called an accident
e Duration of occurrence-the more quickly it happens, the more likely it is to be called an accident
f Degree of negligence-the more reckless or carelessness involved, the less likely it is to be called an accident
g Degree of misjudgment-the more mistakes in judgment involved, the less likely it is to be called an accident
2 An accident is any unplanned and uncontrolled event caused by human, situational, or environmental factors, or any combination of these factors that interrupts the work process, which may or may not result in injury, illness, death, property damage, or other undesired events, but which has a potential to do so (Colling, 1990)
3 An accident is an unplanned and uncontrolled event in which the action or reaction of an object, substance, person, or radiation results in personal injury or probability thereof (Heinrich et al, 1980)
4 Accident is an unplanned, not necessarily injurious or damaging event, which interrupts the completion of an activity, is invariably preceded by an unsafe act and or an unsafe condition or some combination of unsafe acts and/or unsafe conditions (Tarrants, 1980)
Many researchers have proposed theories to explain how accidents occur It is important to understand the theories so that countermeasures can be generated to reduce the occurrences of the accidents These theories also help in undertaking accident preventing actions during the development of safe products and processes This section provides brief descriptions of several accident causation theories that are useful in designing safe products More detailed descriptions of the theories are provided by Petersen and Goodale (1980)
1 Act of God (Demons or other supernatural forces): This theory is recognized in the legal literature and by the insurance industry that some accidents can only be explained as an “act of God” The theory assumes that such acts are bad happenings outside an individual’s control (or reasons such as the victims were presumably marked for punishment because of some unknown quality, the devil did his handiwork, etc)
2 Accidents are “rare and random” events: This theory recognizes that accidents are very low probability events and they can happen and do happen to anyone Here, an accident is considered as a “lottery”—an event whose outcome seems to be determined by chance In early 1900s, the accidents to a large extent were considered uncontrollable and unpredictable Minimum thought was given to design of environments that could reduce probability of an accident or harm It also suggests that repeated violations of commonsense safe practices eventually and invariably will lead to an accident
3 Accident prone theory: This theory assumes that accidents are caused by some invariant human characteristics identified as “accident proneness” Accident proneness may be defined as the continuing or consistent tendency of a person to have accidents as a result of his or her stable characteristics (or response tendencies) Such accident prone people are also called accident “repeaters,” that is, they are involved in repeated number of accidents (or are “over-involved” in accident occurrences as compared to normal individuals) (Note: The accident proneness theory can be tested by analyzing accident data Compare an “observed” distribution of number of individuals involved in x number of accidents in a population with a “fitted” distribution of expected number of individuals involved in x number of accidents
assuming Poisson distribution (P[x], where x is the number of accidents occurring to an individual and x = 0, 1, 2, 3, …) Statistically significant difference between the “observed” and “fitted” (expected) distributions suggests the presence of accident proneness This suggests that individuals with certain accident prone characteristics are “over-involved” in accidents; that is, accident prone individuals have larger values of x than others with smaller values of x)
4 Chain of multiple events: This model assumes the existence of many factors influencing accidents rather than any key cause The probability of an accident (P) that will occur in a given unit of activity (A) is assumed to be a function of a whole set of factors and conditions If these factors are designated as x1, x2, x3, x4, x5, …, xk, the probability of an accident in the activity would be a function of the factors (ie, PA = [x1, x2, x3, x4, x5, …, xk])
5 Energy exchange model: Most accidents are caused by unplanned or unwanted release of excessive amounts of energy (eg, mechanical, electrical, chemical, thermal, and ionizing radiation) or hazardous materials (eg, carbon monoxide, carbon dioxide, hydrogen sulfide, methane, water, and so forth) (However, with a few exceptions, these accidents due to energy releases can also be explained by the unsafe acts theory where an unsafe act may trigger the release of large amounts of energy or a hazardous material, which in turn causes the accident)
6 Epidemiological model: Epidemiology is the study of causal relationship between environmental factors and disease The epidemiological theory suggests that the models used for studying diseases can also be used to study causal relationships between environmental factors and accidents The model assumes that the key components are (a) predisposition characteristics (ie, a susceptible host, a disease producing agent or a virus, and a hazardous environment) and (b) situational characteristics (eg, not wearing sufficiently warm [low insulation] clothing in cold environment, risk taking, and an untrained host) The predisposition characteristics are assumed to create a disease-producing condition or an accident-like condition (ie, unexpected, unavoidable, or unintentional), which in combination with the situational characteristics results in an accident that causes injuries or damage (see Figure 112)
7 Domino theory: An accident is caused by a sequence of events Each event can be assumed to be represented by a domino
The domino sequence is as follows (see Figure 113): a Faults of persons are created by environment or acquired by inheritance b Unsafe acts and conditions are caused by faults of persons
c Accidents are caused by unsafe acts of persons and/or exposure to unsafe conditions
d Injury results from an accident
The Domino theory thus suggests that the above four-events sequence involved in the causation of an accident can be broken by removing any one of the events (like removal of a Domino in the chain) to prevent an accident from occurring
8 Unsafe acts theory: Accidents occur primarily due to unsafe acts of people The unsafe acts occur due to the following reasons: (1) misunderstanding of instructions, (2) lack of knowledge or training, (3) recklessness or violent temper, and/or (4) actions that exceed human capabilities and limitations (eg, speed of response needed was beyond operators capability to react)
9 Human factors or human error models: There are several human factors models that can be related to accident causation The models postulate that human failures occur due to reasons such as (1) task demand exceeds operator capabilities, (2) operator experiences information overload, (3) operator’s attention is diverted/distracted, (4) operator is not consistent in his response (ie, variability in human operator’s output is too high), (5) operator is under stress, and (6) operator fails to get the right information at the right time [or fails to process the information] needed to make the decision and thus does not make the right response needed to avert the accident Most human failures can also be explained as a result of one or more human errors Thus, the human factors models can also be considered to originate from an occurrence of a human error
Many of the human failures are due to information processing errors where the human operator fails to make the right decision at the right point in time, and thus, the accidents are caused by human errors Some examples of human errors are as follows: (1) interpretational errors (ie, errors in interpreting situations or signal interpretation;
for example, the operator misunderstood the meaning of the “red” flashing light on his instrument panel), (2) substitution errors (ie, substituted a different action instead of the intended action; for example, the driver pressed the gas pedal instead of the brake pedal-substituted a wrong control), (3) reversal errors (ie, operator responded with an action in the opposite direction instead of the intended direction; for example, turned a control clockwise instead of counterclockwise), (4) legibility errors (eg, operator could not read a display due to small font size, poorly lit display at night, or sunlight reflection glare from the display lens), (5) forgetting errors or omission errors (ie, forgot to perform an action; for example, pilot took-off without reading the fuel gage), (6) commission errors (ie, performed a task or step when not required or in a different sequence), (7) other errors (ie, control operational errors due to violation of one or more human factors principles in the design of the control; for example, a control is not located at an expected location, the direction of control motion violates its direction-of-motion stereotype, the control not located with other controls of the same functional group, the control was not located close to its associated display, etc)
Two types of measures (ie, variables) are used to measure safety performance They are (1) accident-based measures (ie, based on accident data) and (2) non-accident measures (ie, based on measurements of data from events other than accidents For example, behaviors exhibited during the use of a product, unsafe acts committed, or errors made in operating equipment) The accident-based measures are more believable (or “hard”) as compared with non-accident measures that are regarded as “soft” measures of safety The advantages of the non-accident measures are that they can be obtained without waiting for accidents to occur and the non-accident events occur at much higher rates than accidents that in general occur very rarely On the other hand, the accidentbased measures are “hard” or ultimate measures of safety and have higher face validity
Measurements are essential to determine level of a problem or effects of changes in the design of a product or a system The measurement of safety performance allows us to assess the following types of problems:
1 State of safety, that is, accurately determining the level of safety in an operation or effectiveness of a product and/or process in achieving safety objectives
2 Assist in business planning and safety improvement activities 3 Allow us to evaluate, compare, or calibrate accident prevention initiatives 4 Provide feedback on past safety actions 5 Predict future safety costs
Some examples of currently used accident-based safety performance measures are provided as follows:
1 Number of accidents 2 Number of injury accidents 3 Number of persons injured in accidents 4 Number of disabling injury accidents 5 Number of fatal accidents 6 Number of fatalities in accidents 7 Incident rates (eg, lost time [disabling] injury frequency rate, number of
accidents per 200,000 work hours) 8 Lost work days (number of workdays lost due to accidents) 9 Accident costs
The above measures are computed over a pre-defined exposure (eg, time duration or number of product usage cycles) The accidents can be also categorized by using a number of classification criteria, such as accident type, accident severity, accident location, characteristics of a person involved in the accident, type of equipment, and type of operation or environment involved in the accident
Incident rates are popularly used to measure safety performance in work-related industries (Tarrants, 1980; Goetsch, 2007; NSC, 2012) The commonly used incident rates based on incidents of injuries, illnesses, or fatalities are defined as follows:
N
T =
× IR
( 200,000)
where IR = Total injury and illness incident rate
N = Number of injuries, illnesses and fatalities resulting from the accidents T = Total hours worked by all employees during the period in question 200,000 hours = 100 employees working 40 hours/week times 50 weeks in
1 year
The incident rates based on the 200,000 hours exposure are as follows:
1 Injury rate 2 Illness rate 3 Fatality rate 4 Lost workday cases rate (accident cases where at least one workday was lost
due to an accident) 5 Number of lost workdays rate 6 Specific hazard rate 7 Lost workday injury rate
The problems and issues with the use of the incident rates are as follows:
1 Since they are based on accidents, they are postmortem Thus, one has to wait for accidents to occur before the value of the measure can be computed Thus, long periods of time need to elapse before reliable estimates of the safety measure can be obtained
2 Since accidents are rare events, they are unreliable for small work organizations (ie, they do not accumulate large number of work hours)
3 They are not very useful to predict effectiveness of safety countermeasures in shorter time periods
Some advantages of the currently used accident-based measures are as follows:
1 Quick acceptance (ie, they are an “accepted standard”) as compared with non-accident data, which may be regarded as questionable by many decision makers
2 Motivate management (ie, they get management’s attention and motivate them to take prompt actions)
3 Long history of use 4 Used by government agencies (eg, US Occupational Health and Safety
Administration) and industry associations 5 Easy to calculate 6 Indicate trends in performance 7 Good for self-comparison
Some disadvantages of the currently used accident-based measures are as follows:
1 They can only be computed after the accidents have occurred (ie, they are reactive or postmortem)
2 The numbers can be easily manipulated as many unreported accidents (intentionally or unintentionally) can cause underestimation of the safety problem
3 They may be biased (due to management attitude to restricted work, doctor influence on reporting, worker attitude to light duties, compensation system, motivation to achieve safety awards, and competitions between organizations based on safety performance)
4 The measured number of accidents is typically low making it difficult to establish trends (accidents, in general, are rare events)
5 The accidents differ in severity (ie, the severity of injuries, amount of property damage, and losses are different in different accidents) Thus, comparisons based only on number of accidents can be misleading
6 Some managers or safety specialists may regard an accident as a “once off or a freak” event (Thus, may disregard the accident data)
Unlike the accident-based measures, the nonaccident measures are not standardized by the government agencies or industries However, methods have been used (eg, Behavior Sampling and Critical Incident Technique, see Chapter 16) to measure unsafe acts, unsafe conditions, and errors The frequency and occurrence rates of such incidences have been used to evaluate safety performance (Tarrants, 1980)
Safety costs have been routinely tracked by many organizations They include (1) costs due to accidents and (2) accident prevention costs The costs due to accidents are generally underestimated due to unreported or unaccounted accidents Incidental costs of accidents have been estimated to be four times as great as the actual costs The accident prevention costs include costs of safety analyses, engineering changes, evaluations/tests, reviews, training, protection devices, and so forth (Note: Safety costs are summarized in a later section of this chapter)
The distinction between accident and hazard can be understood by considering the following two considerations:
1 Accident: Accident is an event in which damage to property or injury to personnel has occurred are occurring (ie, accident cases and data are accumulating)
2 Hazard: Hazard is a real or potential condition that could cause damage to property or injury to personnel, but has not occurred so
The following two possibilities need to be considered prior to deciding on type of safety analysis and methods to use (Hammer 1980, 1989; Colling, 1990)
1 An accident has occurred or accidents are occurring: This possibility leads to (a) accident investigation and (b) accident analysis An accident investigation usually precedes an accident analysis
2 Not waiting for accidents to occur: This possibility leads to conducting hazard identification and hazard analysis The hazard analysis involves the following: (a) hazard identification, (b) determining whether controls are in place to prevent occurrences of hazards, (c) formulate countermeasures, and (d) select the best countermeasures to implement to avoid future accidents
Accident analysis can be considered to include the following methods: (1) Accident investigation, (2) accident analysis, and (3) accident data analysis These three methods are not distinctively different and there is considerable overlap between their contents The applications of the methods can also vary depending on the accident
researcher involved in performing the analyses These three methods are described below
1 Accident investigation: The accident investigation involves reading the accident report, visiting the accident site, talking to the witnesses, gathering all facts about a particular accident such as who was involved, how the accident occurred, what the injuries and losses were, and so on A detailed accident investigation typically involves: (a) reading individual accident reports, (b) sending independent accident investigators (or a team of multi-disciplinary experts-in case of detailed investigation) to verify the details of the accident, (c) reconstruct the accident (ie, describe how the chain of events led to the accident), and (d) preparing a detailed report on each accident case
2 Accident analysis: An accident analysis usually involves more than one accident of a given type (eg, accidents involving a particular product model under a certain type of situation, while performing a certain task or a maneuver) The accident analysis involves: (a) collecting and analyzing accident data, (b) determining causes and circumstances of the accidents, (c) investigating possible chains of events that led to the accidents, and (d) creating a model of the accident situation to reconstruct and illustrate details about the behavior of various elements or events that led to the accidents
3 Accident data analysis: The accident data analysis usually involves the following: (a) securing access to one or more accident databases, (b) understanding variables and categories used in creating the database, (c) evaluating completeness of the data (ie, understanding missing data or uncategorized variables that are generally categorized as “other” or “not available”), (d) creating tabular summaries of accident data based on relevant variables of interest, and (e) conducting statistical tests to determine if any of the differences due to variables and their categories are statistically significant
The methods used for accident analysis are described in Chapter 16
The methods used for hazard analysis also have some overlap in content and differences in formats and details depending on the individuals or organizations conducting the analyses The methods used for hazard analysis are listed as follows:
1 General hazard analysis 2 Detailed hazard analysis 3 Methods safety analysis (like operations analysis) 4 Job safety analysis/job hazard analysis (to uncover hazards in a job) 5 FMEA 6 FMECA 7 FTA
8 Error analysis 9 Human reliability analysis
The methods used for hazard analysis are described in Chapter 16 More information on the above listed hazard analysis methods can be obtained from Brown (1976), Hammer (1980), Colling (1990), Roland and Moriarty (1990), and Goetsch (2007)
Product manufacturers are sued in large numbers by users, misusers, and even abusers of their products Injuries resulting from the use (or often misuse) are the basis for an increasing number of product liability lawsuits (Hammer, 1980; Goetsch, 2007) The suits cost the industries millions of dollars each year Therefore, the objectives of this section is to provide the reader background into basic concepts and issues related to the product liability, to understand the role and importance of safety analyses in product and systems design, and to help a product engineer communicate with a product liability lawyer
The best way a manufacturer can prevent or defend such claims is by manufacturing a reasonably safe and reliable product (or system or process), and where necessary by providing instructions for its proper use The key to achieve a reasonably safe and reliable product and to reduce the product liability exposure is to “build-in” product safety during the early design stages of the product
The following terms used in product litigations will help the reader understand the issues (Hammer, 1980; Goetsch, 2007)
1 Liability: It can be defined as an obligation to rectify or to recompense for any injury or damage for which the liable person has been held responsible or for failure of a person to meet a warranty Here, the product user is the loser (or injured) and is assumed to be demanding compensation for the injury, losses, and/or sufferings caused by the product
2 Plaintiff: A person (or a party) who starts a legal case against another to obtain a remedy for an injury caused by the product
3 Defendant: A manufacturer (or the seller) who is faced with proving that the product is safe
4 Three major legal principles: The following three principles are generally considered in establishing liability: (a) negligence, (b) strict liability, and (c) breach of warranty (Hammer, 1980)
a The negligence principle involves failure of the defendant to exercise a reasonable amount of care in the design and manufacture of his product (or to carry out a legal duty) so that injury or property damage does not occur to a user or other person Thus, here the focus is on the conduct of defendant (manufacturer), ie, his duty and/or care The plaintiff must prove that the defendant’s conduct involved: (i) an unreasonably great
risk of causing damage, (ii) defendant failed to exercise ordinary or reasonable care, and/or (iii) not using available knowledge (eg, new developments, design methods, safety practices, and safety devices) that would have decreased the level of risk
b The principle of strict liability as based on the concept that a manufacturer of a product is liable for injury due to a defect, without the necessity for a plaintiff to show negligence or fault Here, the plaintiff must only prove that the product was defective, unreasonably dangerous, and the proximate cause of the harm Thus, in the strict liability, the focus is on the quality of product-rather than the fault of the manufacturer (ie, regardless of whether or not the manufacturer acted reasonably) The manufacturer is said to be “strictly liable” because his liability does not depend on his own conduct or care Therefore, defense is particularly difficult and frustrating to the manufacturer This is the basic cause for what is called the product liability crisis
c The breach of warranty can involve the following two principles: (i) implied warranty and (ii) expressed warranty and misrepresentation The principle of implied warranty involves an implication by a manu-
facturer or dealer that a product is suitable for a specific purpose or use, is in good condition, or is safe by placing it on sale The implied warranty of safety is the principle that any product by being placed on sale is implied to be safe The implied warranty of merchantability implies that the product sold is in as good condition as other products of its type The implied warranty of fitness implies that the product is suitable for the purpose for which it is sold
The principle of expressed warranty involves a statement by a manufacturer or dealer, either in writing or orally, that his product will perform in a specific way, is suitable for a specific purpose or contains specific safeguards
The first step in product liability (in cases where no express warranty or misrepresentation is involved) is to prove that the product was defective (Note: It is not sufficient to establish that the product was dangerous (eg, even a knife can be dangerous) Thus, the product must have a design defect (ie, a defect in its basic design) or a manufacturing defect (eg, a flaw in the manufacturing process)
Some examples of design defects are as follows: (1) a concealed danger created by the design, for example, a sharp edge after collapse in an accident, (2) needed safety devices not included, and (3) involved materials of inadequate strength or failed to comply with accepted standards
Some examples of manufacturing defects are as follows: (1) poor quality material used for structural components, (2) failed to meet required material hardness (eg, failure in the heat treating process), (3) a sharp edge or flash left on a grasp
handle (eg, operator forgot to grind the sharp edge, which can lead to an injury), and (4) improper assembly (eg, misaligned parts, loose parts, missing parts, wrong electrical connections [transposed wires])
The courts have recognized the concept of the product manufacturer’s duty to warn the users about potential safety-related considerations and problems The general position of the courts is to always use warnings The cost of supplying warnings is low as generally only the printing costs are involved (eg, warning labels, warning messages included in screens or user’s manuals) In many cases, the manufacturers have been found to be liable for failure to warn, even when the (missing) warning would have been of dubious value From the Human Factors perspective, too many warnings are ineffective as people will generally disregard frequent occurrences of warnings But courts view provision of warnings as desirable and thus it may have some effect on reducing the manufacturer’s negligence (due to failure to warn) Thus, presence or absence of warnings is generally an issue as compared with their effectiveness
Engineering and management, thus, can be vulnerable in the following areas: (1) product design; (2) product manufacturing and materials selection; (3) packaging, installation, and application/use of the product (ie, operation); and (4) warning labels Failure to comply with the regulations (ie, applicable standards) can mean that the manufacturer is negligent Compliance to government standards (which are generally minimum standards) may provide some protection against negligencerelated cases but it offers no protection in strict liability cases
Safety costs are all costs associated in incorporating safety in the product design and ensuring that the products operate safely during their operational life These costs include the following:
1 Costs incurred in creating safe products: a Costs associated in gathering data on safety regulations, past accidents,
and litigations-related data b Costs associated in developing safety requirements, cascading require-
ments to various systems, subsystems, and components of the product (Note: Safety is a product attribute)
c Costs to design and implement safety features in the product (eg, costs associated in conducting safety analyses, product safety design reviews, meetings with experts, management, government agency experts, lawyers)
d Costs associated in following special safety precautions (eg, checks, inspections) during manufacturing and assembling
e Verification and validation costs (eg, safety testing costs)
2 Costs incurred during product uses/operations a Costs associated in gathering data on safety-related incidences (eg,
meeting with customers, users, dealers, repair shop personnel, government agency personnel, and lawyers; investigating product failures and accidents)
b Conducting safety analyses and tests c Providing technical and legal support on product litigations, recalls,
repairs, fines, customer relations campaigns, and so on d Costs associated with fixing the defects (ie, product recalls, repairs,
retests) 3 Safety costs related to product discontinuation and disposal a Disposal or recycling of retired products b Disposal of plant equipment and hazardous/toxic substances (eg, tox-
icity tests after disposal)
Increased awareness of security issues with computers, databases, and terrorist attacks (eg, cyberattacks) has raised many issues with the security of large complex products (eg, commercial aircrafts, cruise ships, and power plants)
Some important security issues involve: (1) lock out of products from unauthorized users, (2) shielding products from viruses and security threats, (3) resilience from threats (how to make the product performance insensitive to threats), (4) system shutdown or operating under high threat levels, and (5) the abilities of the product to function during and after threat incidences
Product designers, especially with embedded computerized control systems, need to include the security issues along with the safety issues discussed in this chapter
Product safety is an important area and the product designers must make sure that their products are safe Safety requirements must be incorporated very early in the product design as product attribute engineering process Safety reviews must be conducted during both the design and manufacturing phases to ensure that the products do not have any design and manufacturing defects that will increase risks to the users as well as the manufacturer It should be realized that during the product liability cases, the manufacturer is considered to be an expert and very knowledgeable about the product safety requirements, available safety devices, safety technologies, and safety-related design and manufacturing considerations Furthermore, the manufacturer is also expected to provide warnings to the users about any potential hazard associated with the uses of the product The design engineers should maintain proper records on safety analyses and safety-related decisions based on potential benefits of the product to the customers versus costs incurred to make the product reasonably safe for defending their decisions if challenged during any future product reviews and liability cases
