What problems occupy your mind?

[ptgatsby]

A rather simple question or two; What kind of problems do you like to solve? How wrapped up in a problem do you get?

Let me clarify in advance - I'm not talking about what you like learning about (INTPs, please take note), but rather what you find yourself tossing around in your head, trying to solve. Do you try to "solve" if God exists constantly? Do you try to come up with large scale problems like global warming? Do you think about how to fix social problems like crime and poverty? Do you think about interpersonal problems? Do you think about games, about strategies in them? Do you think about hypothetical problems or tangible problems?


Don't answer according to type - really think back over what occupies your mind.

This question brought to you by the realisation that I have been thinking about this "theoretical" problem for nearly two days.
--
You are in charge of a company that does construction. You have a large Health and Safety group and you keep very good track of your accidents and injuries. Over the past 20 years, you have steadily decreased them, up until a few years ago where they stabilized and have begun rising. Over this time you have implemented more and more programs, but the addition of each new program seems to have the opposite effect - an increase in injuries.

You have 250 projects ongoing at this time, with some ending and more being added all the time. Projects operate in all sorts of environments (all over the world) and all forms of construction, however the number of people on the sites are generally the same.

Solve: What approach should be taken in order to find the optimal level of health and safety policies, procedures, training, audits, supervision and so forth. Can you show that there is a maximum threshold of these items after which you have negative returns?

Solve: Once concluding the optimal level of these, how do you determine relative value of each so that new processes can replace old ones if they are found to be more effective? How do you determine they are more effective?

I have been pondering this question while I was trying to sleep. While I was killing demons on my computer. While I try to work, while I eat. I can't get it out of my head.

[Randomnity]

Mostly interpersonal problems, sometimes school/career possibilities and problems, quite often logistics for something I'm planning (studying, social events, activities etc). I'm almost always planning something or other in my head, though often it doesn't happen according to plan.

I usually clear theoretical problems from my mind once I get bored of trying to solve them.

[Athenian200]

Originally Posted by ptgatsby

A rather simple question or two; What kind of problems do you like to solve? How wrapped up in a problem do you get?

Let me clarify in advance - I'm not talking about what you like learning about (INTPs, please take note), but rather what you find yourself tossing around in your head, trying to solve. Do you try to "solve" if God exists constantly? Do you try to come up with large scale problems like global warming? Do you think about how to fix social problems like crime and poverty? Do you think about interpersonal problems? Do you think about games, about strategies in them? Do you think about hypothetical problems or tangible problems?


Don't answer according to type - really think back over what occupies your mind.

I think I usually try to resolve flaws and inconsistencies in things I use. For instance, if something in MBTI theory seems to contradict itself, I'll try to reconcile it, or find a way to discard the part that seems less important. If I have a computer problem, I'll be occupied with trying to find a way to fix it. I'm also typically examining myself and my interaction style to see what I'm doing right, what I'm doing wrong, and trying to improve on it. I also try to form a "picture" of how another person thinks/reacts, and refine it after every interaction, seeing where I guessed correctly and where I didn't.

This question brought to you by the realisation that I have been thinking about this "theoretical" problem for nearly two days.
--
You are in charge of a company that does construction. You have a large Health and Safety group and you keep very good track of your accidents and injuries. Over the past 20 years, you have steadily decreased them, up until a few years ago where they stabilized and have begun rising. Over this time you have implemented more and more programs, but the addition of each new program seems to have the opposite effect - an increase in injuries.

You have 250 projects ongoing at this time, with some ending and more being added all the time. Projects operate in all sorts of environments (all over the world) and all forms of construction, however the number of people on the sites are generally the same.

Solve: What approach should be taken in order to find the optimal level of health and safety policies, procedures, training, audits, supervision and so forth. Can you show that there is a maximum threshold of these items after which you have negative returns?

Solve: Once concluding the optimal level of these, how do you determine relative value of each so that new processes can replace old ones if they are found to be more effective? How do you determine they are more effective?

I have been pondering this question while I was trying to sleep. While I was killing demons on my computer. While I try to work, while I eat. I can't get it out of my head.

Well, I would guess that there is only so much the oversight staff can do, and if they're overburdened, eventually they'll pay too little attention to everything (because they have to divide their attention), and thus will actually be less effective than they were with fewer programs. So you need to figure out how many programs they can handle effectively, and compare the effectiveness of a newer program with that of an older one separately in similar situations. Keep which ever ones are most effective, and discard less effective ones until the number of processes is reduced enough that overall effectiveness rises again.

Did that make sense?

[Uberfuhrer]

Why is there air?

[ptgatsby]

Originally Posted by athenian200

So you need to figure out how many programs they can handle effectively, and compare the effectiveness of a newer program with that of an older one separately in similar situations.

Right, but how? You have a database of 20 years for ~250 construction sites and can track back roughly from 10-100 procedures (over time), some going extinct, being replaced with new ones. New training procedures. New supervision. Hiring of HSQ managers. Different levels of audits.

IOW, probably 15 or so core factors, none of which are consistent. How do you solve the optimal level of each? How do you evaluate the value of new versions (since you can't theoretically add any more on once you reach the optimal level - how do you decide which old less effective ones should be retired?)

[raincrow007]

Currently? The logistics of transporting several large works across several large states, and then [upon arrival] the configuration of said works in a finite, as-yet-unseen-by-me space. I have pages of diagrams, hastily scribbled calculations, lists, and maps of Waffle House locations for a 500 mile radius.

It's about as fanatical and concrete as I get about anything.

[Athenian200]

Originally Posted by ptgatsby

Right, but how? You have a database of 20 years for ~250 construction sites and can track back roughly from 10-100 procedures (over time), some going extinct, being replaced with new ones. New training procedures. New supervision. Hiring of HSQ managers. Different levels of audits.

IOW, probably 15 or so core factors, none of which are consistent. How do you solve the optimal level of each? How do you evaluate the value of new versions (since you can't theoretically add any more on once you reach the optimal level - how do you decide which old less effective ones should be retired?)

You would have to look at the actual data. How much information did the original problem give you? If that's all there was, then you don't have enough information to go that far.

I would guess that if you review all the programs, and find out which ones are redundant, you should eliminate those first. You might also need to separate out similar programs in similar environments, and "test" which ones worked best for a while. You should also try giving similar groups different workloads, and comparing them on that basis, seeing where effectiveness "peaks," and then trying to fit the most effective programs in underneath that optimal/peak workload.

That's assuming that you don't have access to research comparing each of the methods already, however, and then you would only have to experiment with the workloads.

[GZA]

Well... its very type-typical, I guess, so sorry about that, but I've been thinking about my purpose on earth and all that lovely stuff

[Santtu]

I've thought how to make effective propulsion for a cruise missile, even though I dont work in the military industry. Launching methods, engine power needed, etc.. this is a passing thought project. Thought only. (edit: then it doesn't qualify as problem to solve, more of an learning thing. Du'h! It IS hard to turn INTP mode off )

I need to make an article management system for a website, so I've been thinking of how to make the editor for it, and how it should look like.

I've planned how to change everything practical in my life so that I get maximum usability from minimum amount of stuff, with minimum effort in transition from this -> there. This would provide me with ability to switch place of living with a lesser disruption to life, making me more able to take jobs as needed for my career.

I've thought how to speed up my development as a human being. I would keep track on some essential things in my life and follow progress. I've been actively trying to figure out what would be good meters for progress and how so, and how would I notice what affects my development in what.

I've thought about how to solve this low-point of my happiness for my personal relations, and how to help the situation to improve for people that I care about, in their own lives. Many people I know have also experienced a low point in happiness. Whom to trust, where, how.. it's a HUGE human engineering problem. Where are the persons responsible for propagating harmful information? How is my social environment structured? It's a huge puzzle to be solved.

[ptgatsby]

Originally Posted by athenian200

You would have to look at the actual data. How much information did the original problem give you? If that's all there was, then you don't have enough information to go that far.

It's a real problem, not an academic one. I probably have more data than I could count at this point - suffice to say, hundreds of thousands, if not millions of records of just about anything you can think of.

(I can generalise a solution in theory - I can't even begin to flesh out the scope of the problem. It's easy to isolate variables, but the last research I saw identified 19 core variables... the noise in the data is prohibitive to any degree of confidence on which one could justify a proposal to risk reducing security programs. You need a pretty airtight case for that, even for experimenting and control groups.)