What Are the Best Practices for Designing Resilient and Adaptive Systems?

Click For Summary

Discussion Overview

The discussion revolves around best practices for designing resilient and adaptive systems in the context of product liability and safety. Participants explore the challenges of anticipating all possible scenarios in system design, particularly under financial constraints, and consider various approaches to improve resilience and adaptability in their designs.

Discussion Character

  • Debate/contested
  • Technical explanation
  • Conceptual clarification

Main Points Raised

  • One participant highlights the difficulty of anticipating every possible scenario in system design due to financial constraints and questions if there are better paradigms for resilience and adaptability.
  • Another suggests hiring radical thinkers or using 'fuzzy logic' predictive software to envision more possibilities in design.
  • A participant emphasizes the fear surrounding product liability and outlines the designer's responsibility to consider all potential uses of a product, regardless of how unlikely they may seem.
  • Concerns are raised about the implications of human behavior on system design, noting that users often make modifications that can lead to unforeseen issues, especially when software updates are involved.
  • There is a call for better internal methods and processes to alleviate the burden on designers and improve overall design practices.
  • Discussion includes the need for comprehensive documentation to clarify the limitations of products and protect the company from liability.

Areas of Agreement / Disagreement

Participants express a range of views on the challenges of product design and liability, with no clear consensus on the best practices or processes to adopt. The discussion remains unresolved regarding the most effective strategies for enhancing resilience and adaptability in system design.

Contextual Notes

Participants acknowledge the limitations of their internal methods and the unpredictability of user behavior, which complicates the design process. There is also mention of varying responsibilities based on the type of product being designed, indicating that context may influence the approach to resilience and safety.

misgfool
I work in an company making moderately complicated instruments or systems. For most cases they work within normal parameters. However, it seems that there are some situations that have not been anticipated. The design in our company sets barriers in front of hazards that may cause accidents. This includes using design standards, best practices, extensive documentation and testing etc. Apparantly it is still nearly impossible to conceive every possible scenario considering the financial constraints. So is there a better paradigm for design and production for systems to be as resilient and adaptive as possible to different conditions?
 
Engineering news on Phys.org
That seems a bit vague; I'm not sure exactly what you want to achieve. If I understand you correctly, all that I can think of would be to hire some radical thinkers or use 'fuzzy logic' predictive software to envision more possibilities than the normal crew comes up with.
 
Danger, what is being asked about is product liability.

I must say that those two words invoke a lot of fear in many industries. I remember sitting through a training seminar given by our head lawyer and a consultant. It painted a pretty poor picture from a designer's standpoint. The things I pulled from it were:

- It is the designer's job to do whatever it takes to think of every possible use your product could be used for, right or wrong.
- Don't e-mail questionable or harmful data, even within your company.
- No matter what, some moron will be out there that will get hurt using your product because you can't design for everything.

I wish I could say that there is a process. Other than your company having it's own internal methods and following as many established codes. I think you need to present anything you can think of and then let your management make the decision as to whether or not to protect themselves by designing into the product a means to prevent each scenario.
 
Thanks for the clarification, Fred, and apologies to Misgfool for failing to understand.
This can indeed become ridiculous after a while. It's sort of like someone who originally just disapproves of trophy hunting escalating to becoming a vegan. If every effort is taken to ensure that there's no possibility of something being misused, nothing could ever work at all. We'd have to start making steak knives out of cotton. No, wait... that could be used to choke someone... :rolleyes:
 
FredGarvin said:
- Don't e-mail questionable or harmful data, even within your company.

What is questionable or harmful data?

FredGarvin said:
- No matter what, some moron will be out there that will get hurt using your product because you can't design for everything.

That's why I have to try to develop better ways of doing things. Human users are not always morons, but when they learn to use the instruments they start making tweaks in their environment. At the same time we are updating for example the software. Now we don't know of the tweaks the customer is making and customer may not fully understand how our updates are affecting operation of the instruments. So eventually disaster strikes.

FredGarvin said:
I wish I could say that there is a process. Other than your company having it's own internal methods and following as many established codes. I think you need to present anything you can think of and then let your management make the decision as to whether or not to protect themselves by designing into the product a means to prevent each scenario.

Don't the internal methods form the process? Trying to improve it takes some of the burden from the poor designer. I would like to hear/read about internal methods and way of thinking in your or any other company.
 
Danger said:
Thanks for the clarification, Fred, and apologies to Misgfool for failing to understand.

No problem. But could I ask, that you would not use a capital letter in the beginning of misgfool. It's not my (birth)name.

Danger said:
This can indeed become ridiculous after a while. It's sort of like someone who originally just disapproves of trophy hunting escalating to becoming a vegan. If every effort is taken to ensure that there's no possibility of something being misused, nothing could ever work at all. We'd have to start making steak knives out of cotton. No, wait... that could be used to choke someone... :rolleyes:

This must be the fuzzy logic part?
 
misgfool said:
No problem. But could I ask, that you would not use a capital letter in the beginning of misgfool. It's not my (birth)name.
For sure. Sorry. I just automatically capitalize usernames because they're proper nouns. I'll refrain from doing so with yours in the future.

misgfool said:
This must be the fuzzy logic part?

:smile:
Absolutely. Lint everywhere. :biggrin:
 
misgfool said:
What is questionable or harmful data?
By that I mean any kind of test data, opinions, etc...that can be pulled out in a trial to show what your company may or may not have known about the liability and why you made certain decisions.

misgfool said:
That's why I have to try to develop better ways of doing things. Human users are not always morons, but when they learn to use the instruments they start making tweaks in their environment. At the same time we are updating for example the software. Now we don't know of the tweaks the customer is making and customer may not fully understand how our updates are affecting operation of the instruments. So eventually disaster strikes.
That is where you need the best documentation stating the limitations that your company expects the customers to in terms of their tweaking. If they go beyond those limits you may have covered your butts...but maybe not.

misgfool said:
Don't the internal methods form the process? Trying to improve it takes some of the burden from the poor designer. I would like to hear/read about internal methods and way of thinking in your or any other company.
In my line of work, our company doesn't really have to worry about what you are asking for. We have different headaches in proving that our engines won't break down, etc...However, if we were designing lawnmowers, you bet we would have to do everything imaginable to prevent someone from hurting themselves. Those kinds of things are the responsibilities of the designers.
 

Similar threads

  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 11 ·
Replies
11
Views
5K
  • · Replies 2 ·
Replies
2
Views
7K
Replies
29
Views
6K
  • · Replies 1 ·
Replies
1
Views
4K
  • · Replies 2 ·
Replies
2
Views
11K
  • · Replies 3 ·
Replies
3
Views
750
  • · Replies 62 ·
3
Replies
62
Views
12K
  • · Replies 2 ·
Replies
2
Views
8K