What are the problem solving skills you gain from.... (Physics vs. Engineering)

[zachdr1]

How are the problem solving skills you gain from majoring in Physics and majoring in Engineering different?

I have noticed that a lot of people think you get better at problem solving by majoring in Physics when compared to engineering, and I don't understand why. I mean, they both study the same things, Physics is just more research based.

I understand that in engineering, you don't have to worry about the proofs, and you don't have to understand how everything comes from the basic principles to get the degree, but most of the time, the people that skip out on learning these things do not get high GPA's.

I make sure I read the textbooks for all of my engineering classes thoroughly, and it seems that they all explain the proofs behind the topics taught in engineering, and don't just present laws without proof along side a huge "trust me" button that every student has to press.

Can anyone else help me explain how the problem solving you do is different in physics vs. engineering?

 

[Dr.D]

There is a difference of emphasis. Physics is more focused on knowledge, while engineering is more focused on accomplishing a particular goal. The physicist is typically much more concerned about idealized cases while engineering deals more with actual cases.

Consider for example how both of them look at a slider-crank mechanism as used in an IC engine. The physicist usually looks at conditions at Top Dead Center (TDC), at Bottom Dead Center (BDC), or the two positions where the crank and the connecting rod are mutually perpendicular. The engineer, on the other hand must deal with every position around the full crank revolution. Similarly, the engineer must deal with piston offset, and an asymmetrical connecting rod, things I have never yet see physicists consider.

The difference in a nutshell is that engineers must deal with the real case while physicist focus on the idealized case. Thus, for example, the physicists interest in cosmology while the engineers says, "we cannot change any of that, so it is not of interest."

 

[Logical Dog]

I mean, they both study the same things

No!

 

[zachdr1]

There is a difference of emphasis. Physics is more focused on knowledge, while engineering is more focused on accomplishing a particular goal. The physicist is typically much more concerned about idealized cases while engineering deals more with actual cases.

Consider for example how both of them look at a slider-crank mechanism as used in an IC engine. The physicist usually looks at conditions at Top Dead Center (TDC), at Bottom Dead Center (BDC), or the two positions where the crank and the connecting rod are mutually perpendicular. The engineer, on the other hand must deal with every position around the full crank revolution. Similarly, the engineer must deal with piston offset, and an asymmetrical connecting rod, things I have never yet see physicists consider.

The difference in a nutshell is that engineers must deal with the real case while physicist focus on the idealized case. Thus, for example, the physicists interest in cosmology while the engineers says, "we cannot change any of that, so it is not of interest."

This is true, but how does that change the kind of problem solving skills you would gain? It seems to me that they would be the same as long as an engineering student doesn't skip learning the theory and relate everything to the basics.

 

[zachdr1]

No!

They both study topics in physics is what I mean.

 

[Dr.D]

To continue with my IC engine example (which is in many ways typical of the whole discussion), the physicist never develops practical ways to deal with the detailed kinematics (including piston offset and con rod asymmetry), and is thus limited to the idealized and simple cases. If you want to deal with the real geometry, including non-ideal geometry, practical ways of dealing with the actual geometry must be addressed. I've never yet seen a physicist do this.

 

[zachdr1]

To continue with my IC engine example (which is in many ways typical of the whole discussion), the physicist never develops practical ways to deal with the detailed kinematics (including piston offset and con rod asymmetry), and is thus limited to the idealized and simple cases. If you want to deal with the real geometry, including non-ideal geometry, practical ways of dealing with the actual geometry must be addressed. I've never yet seen a physicist do this.

Ahhh that's true. This seems like the only difference though.

 

[Dr.D]

It is a difference with very big significance if you want to study the details of IC engine operation and vibrations. If those topics are of no interest (as they are for most physicists), then the difference does not matter.

There is an apocryphal story that may shed some light here.

A mathematician, a physicist, and an engineer are all attending a meeting in a hotel. The smell smoke and come out of their hotel rooms to discover the curtain at the end of the hall is own fire.

The mathematician arrives first, observes that there is a near by fire extinguisher, so he is satisfied that the problem has a solution and he leaves.

The physicist arrives second, and hastily makes an estimate of the total heat available from the burning curtain and the total heat that can be removed by the fire extinguisher. Having determined that the fire extinguisher is adequate to put the fire out, he too leaves.

The engineer arrives last, grabs the fire extinguisher and puts the fire out. Then he leaves.

Do you see any difference in their responses?

 

[rwm4768]

I would say Engineering actually teaches you more in terms of problem solving skills, especially when you get into upper level classes, where you frequently have to solve problems in some design you're working on. Not that physics classes don't teach these skills themselves. They do to some extent. But I think the focus of physics courses is to understand physics while the focus of engineering courses is to understand physics so you can use it to solve problems.

Obviously, there's some overlap, especially if you go into applied physics.

 

[micromass]

It is a difference with very big significance if you want to study the details of IC engine operation and vibrations. If those topics are of no interest (as they are for most physicists), then the difference does not matter.

There is an apocryphal story that may shed some light here.

A mathematician, a physicist, and an engineer are all attending a meeting in a hotel. The smell smoke and come out of their hotel rooms to discover the curtain at the end of the hall is own fire.

The mathematician arrives first, observes that there is a near by fire extinguisher, so he is satisfied that the problem has a solution and he leaves.

The physicist arrives second, and hastily makes an estimate of the total heat available from the burning curtain and the total heat that can be removed by the fire extinguisher. Having determined that the fire extinguisher is adequate to put the fire out, he too leaves.

The engineer arrives last, grabs the fire extinguisher and puts the fire out. Then he leaves.

Do you see any difference in their responses?

A mathematician, a physicist and an engineer are required to measure the volume of a red ball.

The mathematician accurately measures the circumference and uses formulas to find the volume.

The physicist dips the ball in water and measures the volume that way.

The engineer looks up the volume in a book by the company that made the balls.

 

[micromass]

A mathematician, a physicist and an engineer are required to measure the volume of a red ball.

The mathematician accurately measures the circumference and uses formulas to find the volume.

The physicist dips the ball in water and measures the volume that way.

The engineer looks up the volume in a book by the company that made the balls.

The philosopher comes back after 10 days and proudly exclaims that he now found out that the ball isn't really there and is an illusion of our senses.

Sorry, couldn't resist...

 

[Dr. Courtney]

I've noticed that physicists are more keen on conducting tests and experiments needed to gather information for solving a practical problem.

Engineers tend to be more trusting of specs and stuff looked up in books or marketing materials. They are more likely to look for problems with how the pieces are put together rather than with the pieces themselves. They are more likely to believe the equations.

On the whole, I like working in teams where both physicists and engineers are well represented.

 

[FactChecker]

Engineers can make a problem as simple or complicated as they want. They are building a system using techniques that they are familiar with. In physics, the complexity is what it is. You either figure it out or you don't.

 

[Dr.D]

I've noticed that physicists are more keen on conducting tests and experiments needed to gather information for solving a practical problem.

Engineers tend to be more trusting of specs and stuff looked up in books or marketing materials. They are more likely to look for problems with how the pieces are put together rather than with the pieces themselves. They are more likely to believe the equations.

On the whole, I like working in teams where both physicists and engineers are well represented.

There is a side to this that may not be familiar to you. In many engineering situations, by law, the drawings and specifications are legally binding. The implication of this (when it applies) is that everyone is required to operate on the assumption that these documents are true and correct. Thus, for example, a piece of 1020 CR steel must have 20 points of carbon, not more, not less (within prescribed allowable errors). If the drawing calls for 1020 CR, then there is no reason to question the carbon content (unless there is a subsequent failure, at which points all bets are off).

In the event of failure analysis (an area where I spent several years), testing is often done by engineers, provided there is suitable material to test. After an accident, fire, water, or other effects often have changed properties to the point that testing is no longer useful.

As to the matter of being more likely to believe the equations, that is probably true. And the equations are probably true (the derivations are very carefully done), but there is always the open question as to whether the correct equations have been properly applied. This is a difficult matter in that, without the equations, all design would have to be done empirically (build - test - build again - test again - ...), but the equations are often not completely applicable. I'm thinking, for example, of the case of a welded frame. What equations do we use? If we treat the frame as pin jointed, we can get a fairly simple solution in most cases, but we know that the welds are not pin joints. If we insist on correctly modeling the welded joints, we almost always move immediately to the need for a finite element model for even the simplest frame.

Dr. Courtney brings up some very good points.

 

[Dr.D]

Engineers can make a problem as simple or complicated as they want. They are building a system using techniques that they are familiar with. In physics, the complexity is what it is. You either figure it out or you don't.

I think that FactChecker has misstated the situation a bit, on two accounts.
1. "They are building a system using techniques that they are familiar with." The implication seems to be that it has all been done before, but this simply is not true. Think about the designers of the equipment for the moon probes. None of this had been done before, and yet, engineers were called upon to build systems that would work the first time - no repeats accepted. Obviously, this was not 100% achieved, but it was most certainly the goal.
2. In the discussion of the IC engine problem that I used above, I said

If you want to deal with the real geometry, including non-ideal geometry, practical ways of dealing with the actual geometry must be addressed. I've never yet seen a physicist do this.

I see this as a clear example of the physicists declining to address the complexity, usually on the grounds that the problem is not interesting (even though many of them drive cars and run lawn mowers).

 

[Dr. Courtney]

There is a side to this that may not be familiar to you. In many engineering situations, by law, the drawings and specifications are legally binding.

I can think of a lot of counter examples, from electronics specifications to ballistics. Had a conversation with an attorney a few years ago contemplating a suit against a manufacturer for consistently failing to meet their ballistics specifications. The proof is absolute and compelling. There has yet to be a lawsuit.

On the other hand, auto manufacturers get sued all the time, not for failing to meet their specs, but for not designing cars that the nation's biggest idiots cannot manage to kill and maim themselves in.

As to the matter of being more likely to believe the equations, that is probably true. And the equations are probably true (the derivations are very carefully done), but there is always the open question as to whether the correct equations have been properly applied. This is a difficult matter in that, without the equations, all design would have to be done empirically (build - test - build again - test again - ...), but the equations are often not completely applicable.

Again, I can think of a lot of counter examples - engineers using equations that are in the books but really have not been validated in any applicable situation remotely similar to how the engineers are using it. But I'm a physicist. I doubt every equation until I've reviewed the experimental evidence supporting its use.

 

[FactChecker]

I think that FactChecker has misstated the situation a bit, on two accounts.
1. "They are building a system using techniques that they are familiar with." The implication seems to be that it has all been done before, but this simply is not true. Think about the designers of the equipment for the moon probes. None of this had been done before, and yet, engineers were called upon to build systems that would work the first time - no repeats accepted. Obviously, this was not 100% achieved, but it was most certainly the goal.

Perhaps I should state my point more carefully. Engineers are using man-made techniques to build man-made systems. They are in control of what they design and naturally limit the complexity to something that engineers can deal with. That is not true when it comes to understanding phenomena like physics, health, economics, climate change, etc. They are as complex as nature wants to make them. For the most part, we still don't understand it all.

 

[Dr.D]

I can think of a lot of counter examples, from electronics specifications to ballistics. Had a conversation with an attorney a few years ago contemplating a suit against a manufacturer for consistently failing to meet their ballistics specifications. The proof is absolute and compelling. There has yet to be a lawsuit.

On the other hand, auto manufacturers get sued all the time, not for failing to meet their specs, but for not designing cars that the nation's biggest idiots cannot manage to kill and maim themselves in.

It is interesting that you bring up the ballistics example; many years ago, I worked in the time fuze industry, and went through a law suit that ended up bankrupting my employer. A ballistics situation is one in which there certainly is definitive proof of function or failure to function, but I'm not sure that is a counter example. What does a consistent failure show? In my experience, it means that the design is inadequate, but contracts are let based on designs (usually done by the government) to build for a fixed price. This means that a manufacture must use the full range of allowable tolerance to keep costs down, but with an inadequate design, that can result in too many failures. The buyer/designer wants to get the price down, so they are pushed to allow generous tolerances, but that may make it impossible to meet the performance requirements.

Your example of the suits against auto manufacturers also does not show me a counter example. This is not a failure of engineers or engineering methods, but rather is a reflection of our insane legal system and greedy lawyers.

Again, I can think of a lot of counter examples - engineers using equations that are in the books but really have not been validated in any applicable situation remotely similar to how the engineers are using it. But I'm a physicist. I doubt every equation until I've reviewed the experimental evidence supporting its use.

The "ability" to misapply existing equations is neither novel nor rare. It is a category of human error. Engineers make human errors, and I suspect that physicist do as well (although I do not have the empirical proof of that last). However, to demand experimental evidence for every part of a system, even before any parts have been made, is unrealistic. At the design stage, nothing exist other than ideas on paper or in CAD. If you want to make any predictions at all, you must use a formulation that exists. It is usually most safe to start at the most fundamental level (F = m*a, etc), rather than with some canned result. Would you still insist on verifying the applicability of F = m*a? And again, how is this a counter example to what I said? I fail to see it.

 

[mpresic]

I have taken upward of a dozen physics and electrical or aerospace engineering graduate courses. I feel physics problem sets were trickier and required greater breadth across the subject. I think it was easier to treat engineering problem sets systematically.

Both physics and engineering problem sets require considerable effort. Usually the physics problem sets required more differential equations than EE, although the EE sets in control theory required more (applied) linear algebra than the physics sets.

 

[Mmm_Pasta]

There is more to problem solving than solving relatively simple problems in a textbook. Engineering curriculum usually involve a senior design project where students have to work with team members and propose some design. They need to take into consideration things such as budget, time, and how practical the design is. At my university they got feedback from those with actual experience in industry. I did physics and math in undergrad so I do not know the entire process, but my roommate was an engineering student, so I did get some sort of idea when I talked to him and his friends.

A traditional physics curriculum usually involves solving simple theoretical problems. I don't really see much in actually working with a team and working on a project as engineering students do. You have to work with other people and there will be problems associated with this. These sorts of problems aren't something you gain experience in solving through a textbook. What if you have to work with budget and time constraints? There are problems associated with this.

Sure, both physics and engineering students work on simple textbook problems, but engineering students have more experience working with teams and dealing with problems that can arise while working on a team on some project that has constraints. It may not be a lot of experience, but certainly more experience than someone who goes through a traditional physics curriculum. When someone with a physics degree mentions "excellent problem solving skills" I just wonder what kind of problems they actually have experience in solving.

 

[symbolipoint]

It may not be a lot of experience, but certainly more experience than someone who goes through a traditional physics curriculum. When someone with a physics degree mentions "excellent problem solving skills" I just wonder what kind of problems they actually have experience in solving.

?
!
!

 

[Mmm_Pasta]

?
!
!

I was being unfair. There are quantitative skills and you do develop some computational skills. Although, I was answering the OP's question in terms of physics vs. engineering and I think there are other problem solving skills that engineers develop.

 

[mpresic]

I agree with many points made by Mmm Pasta. The last two engineering courses I took, (the only ones I took in the last 12 years) required us to work in teams. In addition, I think the professor did this deliberately, because where I (and I presume others) might call an end to an open ended assignment early, no one wants to let the team down. Actually one of my grad physics project was also done in teams. I remember pulling an all-nighter before the presentations, because my team mates were also committed. I remember I would never have gone to this extreme had I been doing the assignment alone.

I object to the characterization of textbook problems as being simple, though. I found many textbook problems are extremely challenging. I remember an abstract algebra textbook that came with a warning that even the author was not sure every problem listed was doable with the information presented so far in the text where the problem was posed.

The team approach simulates real life to a large degree. Most problems at work are solved by a team. On the other hand, evaluations (i.e. grades) are presumed to be given on an individual basis. When I am on a hiring committee looking for engineers, I would be comforted to know the grades are reflective of the performance of the individual I would be considering, and not the whole "team" which I would not be hiring.

I feel physics made me a better problem solver than engineering, with the following caveats. I learned physics first, so that I was already a good problem solver before I took engineering courses. It could be if I learned engineering first, I might feel that engineering made me a better problem solver and prepared me for physics.

I am also open to the suggestion possibly given by mathematicians, that "both you guys/women got it wrong. Mathematics provides the best problems which require the creative approach. If you can do math problems, the physics and engineering problems come easily".