**5. Decision tools**

Deciding on whether to introduce a new technology is a choice made under risk and uncertainty. This is why factors of safety are a part of every engineering recommendation. A number of decision support tools are available to aid in predicting the ethical implications of a new technology. Most are at best semi-quantitative. Their major strength lies in their objective descriptions needed for selecting among various alternatives. They even allow for some degree of weighting among physical and social values.

#### **5.1 Net goodness analysis and decision trees**

The net goodness analysis aids in the previously mentioned good works model. It estimates the goodness or wrongness of an action by weighing its morality, likelihood, and importance. This is a subjective analysis of whether a decision will be moral or less than moral. It puts the case into perspective, by looking at each factor driving a decision from three perspectives: 1. how good or bad would the consequence be; 2. How important is decision; and 3. how likely is it that the consequence would occur. These factors are then summed to give the overall net goodness of the decision:

considerations." Public safety and health considerations affect the design process directly. Almost every design now requires at least some attention to sustainability and environmental impacts. For example, recent changes in drug delivery have been required, such as moving away from the use of greenhouse gas propellants like chlorofluorocarbons (CFCs) and instead using pressure differential systems (such as physical pumps) to deliver medicines. This may seem like a small thing or even a nuisance to those who have to use them, but it reflects an appreciation for the importance of incremental effects. One inhaler does little to affect the ozone layer or threaten the global climate, but millions of inhalers can produce enough halogenated and other compounds that the threat must be considered in

Technologists must consider how sustainable the technology will be over its useful lifetime. This requires thinking about the life cycle, not only during use, but when the use is complete. Such programs as "design for recycling" (DFR) and "design for disassembly" (DFD) allow the engineer to consider the consequences of various design options in space and time. They also help designers to pilot new systems and to consider scale effects when ramping up to full production of devices. However, if such a change inordinately affects a vulnerable population, this must be weighted properly in the decision. For example, if asthmatics are placed at additional risk due to a less efficacious delivery system, albeit more environmentally acceptable, it is likely not the best alternative. That is, the risk tradeoff between biomedical and environmental values leans more heavily toward the biomedical

This illustrates that like virtually everything else in engineering, best serving the public is a matter of optimization. The variables that we choose to give large weights will often drive the design. The technologist must continue to advance the state-of-the science in high priority areas. Any possible adverse effects must also be recognized. These should be incorporated and properly weighted when we optimize benefits. We must weigh these

Deciding on whether to introduce a new technology is a choice made under risk and uncertainty. This is why factors of safety are a part of every engineering recommendation. A number of decision support tools are available to aid in predicting the ethical implications of a new technology. Most are at best semi-quantitative. Their major strength lies in their objective descriptions needed for selecting among various alternatives. They even allow for

The net goodness analysis aids in the previously mentioned good works model. It estimates the goodness or wrongness of an action by weighing its morality, likelihood, and importance. This is a subjective analysis of whether a decision will be moral or less than moral. It puts the case into perspective, by looking at each factor driving a decision from three perspectives: 1. how good or bad would the consequence be; 2. How important is decision; and 3. how likely is it that the consequence would occur. These factors are then

designing medical devices.

value (treating asthma effectively).

**5. Decision tools** 

benefits against possible hazards and societal costs.

some degree of weighting among physical and social values.

summed to give the overall net goodness of the decision:

**5.1 Net goodness analysis and decision trees** 

#### NG = (goodness of each consequence) x (importance) x (likelihood) (3)

These analyses sometimes use ordinal scales, such as 0 through 3, where 0 is nonexistence (e.g. zero likelihood or zero importance) and 1, 2 and 3 are low, medium and high, respectively. Thus, there may be many small consequences that are near zero in importance and, since NG is a product, the overall net goodness of the decision is driven almost entirely by one or a few important and likely consequences. There are two cautions in using this approach. First, although it appears to be quantitative, the approach is very subjective. Second, as we have seen many times in cases involving health and safety, even a very unlikely but negative consequence is unacceptable.

The tool can be modified from a purely ethical decision making tool to a risk management tool by incorporating the net goodness into a decision tree. For example, Fig. 8 shows a hypothetical decision on whether to use a GMO (Vallero 2010). The decision is based on the likelihood of various beneficial and adverse outcomes, with ranked importance to three receptors: the environment; public health; and food production. The analysis is qualitative, but can help to identify important factors, as well as potential downstream impacts and artifacts of an immediate decision. The difficulty will be to arrive at probabilities to fill the "likelihood" column. Sometimes these are published, but often will have to be derived from focus groups and expert elicitation. Often, likelihood is presented as an ordinal scale (e.g. high, medium, or low – or 1, 2, or 3).


#### **1 = Best; 5 = Worst**

\*This has its own decision tree according to vulnerability index, i.e. percentile exposure (high to no exposure) and sensitive subpopulations (children, elderly, asthmatic, etc.)

Fig. 8. Decision tree and net goodness analysis of a decision to insert *Bacillus thuringiensis*  genetic material into crops near an ecosystem. Data are hypothetical. Vallero 2010).

#### **5.2 Line drawing**

Line drawing (Fledderman 2011) is most useful when there is little disagreement on which moral principles apply, but when there is no consensus about how to apply them. The approach allows the comparison of several real-world precedents for which there is general agreement about right and wrong. The emerging technology (the unknown) is plotted for each important factor (e.g. safety, privacy, etc.) Two of the precedents are extreme cases of right and wrong, respectively. The positive paradigm is very close to being unambiguously moral and the negative paradigm is unambiguously immoral:


Next, the emerging technology (T) is put on a scale showing the positive paradigm (PP) and the negative paradigm (NP), as well as other cases that are generally agreed to be less positive than PP but more positive than NP. This shows the relative position of T:

Ethical Decisions in Emergent Science, Engineering and Technologies 45

collectively. Fleddermann (2011) has used a flow chart for the toxic gas release at Bhopal, India. This flow chart addresses one specific decision, i.e. where to site the plant. Other charts need to be developed for safety training, the need for fail-safe measures, and proper operation and maintenance. Thus, a "master flow chart" can be developed for all of the decisions and sub-consequences that ultimately led to the disaster. A similar chart can used to consider possible contingencies and decision points for an emerging technology, such as

Fig. 9. Flow chart on whether to use a genetically modified organism (GMO) for an

environmental cleanup (Vallero 2010).

the one in Fig. 9 regarding the decision to use a GMO.

This gives the sense that the new technology is more positive than negative, but still short of being unambiguously positive. In fact, two precedents (2 and 3) are much more morally acceptable. This may indicate the need to consider taking an approach similar to these precedents, at least for the most sensitive factors (those that have influenced the location on the line).

#### **5.3 Charting**

Critical paths, PERT charts and other flow charts are useful in ethical analysis if sequences and contingencies are involved in reaching a decision, or if a series of events and ethical and factual decisions lead to the consequence of interest. Thus, each consequence and the decisions that were made along the way can be seen and analyzed individually and

Line drawing (Fledderman 2011) is most useful when there is little disagreement on which moral principles apply, but when there is no consensus about how to apply them. The approach allows the comparison of several real-world precedents for which there is general agreement about right and wrong. The emerging technology (the unknown) is plotted for each important factor (e.g. safety, privacy, etc.) Two of the precedents are extreme cases of right and wrong, respectively. The positive paradigm is very close to being unambiguously

Next, the emerging technology (T) is put on a scale showing the positive paradigm (PP) and the negative paradigm (NP), as well as other cases that are generally agreed to be less

This gives the sense that the new technology is more positive than negative, but still short of being unambiguously positive. In fact, two precedents (2 and 3) are much more morally acceptable. This may indicate the need to consider taking an approach similar to these precedents, at least for the most sensitive factors (those that have influenced the location on

Critical paths, PERT charts and other flow charts are useful in ethical analysis if sequences and contingencies are involved in reaching a decision, or if a series of events and ethical and factual decisions lead to the consequence of interest. Thus, each consequence and the decisions that were made along the way can be seen and analyzed individually and

positive than PP but more positive than NP. This shows the relative position of T:

moral and the negative paradigm is unambiguously immoral:

**5.2 Line drawing** 

the line).

**5.3 Charting** 

collectively. Fleddermann (2011) has used a flow chart for the toxic gas release at Bhopal, India. This flow chart addresses one specific decision, i.e. where to site the plant. Other charts need to be developed for safety training, the need for fail-safe measures, and proper operation and maintenance. Thus, a "master flow chart" can be developed for all of the decisions and sub-consequences that ultimately led to the disaster. A similar chart can used to consider possible contingencies and decision points for an emerging technology, such as the one in Fig. 9 regarding the decision to use a GMO.

Fig. 9. Flow chart on whether to use a genetically modified organism (GMO) for an environmental cleanup (Vallero 2010).

Ethical Decisions in Emergent Science, Engineering and Technologies 47

Researchers over the past two decades have embarked on ways to ensure that their graduate and faculty research not only advances the state of the science, but do so with integrity. Comparing decision making in emerging areas like these, especially nanomaterials and biotechnologies, to that of more established scientific enterprises combines technical and ethical content of decisions to go forward. As mentioned, Responsible Conduct of Research (RCR) programs at universities include training on ethics topics in specific research areas using proven educational resources, but newer techniques are required when dealing with

Indeed, RCR has been a key part of practical training of research and teaching assistants, provides a bridge between professional and research ethics, helps to ensure transparency and documentation of funding, and is vital to preparing the next generation of scholars by promoting research that both gains the public trust and contributes to the needs of society. As such, RCR provides a means of "preparing stewards of the discipline" as posited by the Carnegie Initiative on the Doctorate. (Duke University 2012). However, research enterprises must to ensure that graduate-level researchers in emerging fields are adequately prepared when confronted with ethical issues associated with emerging technologies (NAS 2004).

RCR training and the professional codes of ethics are starting points for ways to approach

Even a well-conceived, thoughtfully designed and carefully deployed technology can fail to meet ethical standards. For example, a noble outcome does not justify unethical means. Failure to follow appropriate safety protocols (e.g. physical containment in a biotechnology) or or violating research norms (plagiarizing or fabricating data) would be an unethical act. Even if standard protocols are followed, this is not sufficient if they do not properly apply to

Ethical decisions must embody systems thinking and consideration of worst case scenarios. This goes beyond obvious misuse and abuse. More subtle drawbacks and abuses need to be avoided by researchers and practitioners. Due diligence requires that one considers all possible good and bad outcomes of an emerging technology. Good practice requires that even a good technology needs commensurate safety and security measures to ensure that it is not misused, since emerging technologies have few, if any, completely reliable precedents.

http://files.asme.org/asmeorg/NewsPublicPolicy/Newsletters/METoday/article

Bronowski, J. (1958). *Science and Human Values*, Harper and Row, ISBN: 0060972815, New

Department of the Environment, United Kingdom Government (1994). *Sustainable* 

American Society of Mechanical Engineers – ASME (2011). *Engineering Ethics.* 

*Development, the UK Strategy,* Cmnd 2426, HMSO, London, UK.

s/28207.pdf*;* accessed on January 11, 2012.

**6. Responsible conduct of research** 

emerging technologies.

an emerging technology.

**8. References** 

York, NY.

**7. Conclusions** 

complex systems (National Academy of Engineering 2003).

Event trees or fault trees also support optimization of factors when weighing alternative versions of a technology or ways to mitigate potential adverse effects. This can help to avoid a technology's feature that can cause harm or lead to failures. For example, Fig. 10 shows an event tree that likely will support using a safer material when fabricating a device.

Fig. 10. Event tree on whether to use mercury in a medical device.
