design,production: new product

[yudansha]

good day to all I am a student of mechanical ing,
I wanted to open a general discussion about a "problem" regarding the production of a new product.
reducing the stages to the bone, we could say:
the design department delivers to the production the engineering of the product (models cad, etc.). at this point the production department defines the various technological cycles, times and methods to send the new product into production (magari on a pilot line).
before the product enters production line in a definitive way, it often returns to the design stage to solve problems that occur in the production phase.

what I ask at this point, referring to your practical experience on the field:
What are the most recurring motivations of these "loop-backs"?

in an optimisation of times, I wonder (always with the support of those who live these things every day): Is there any information that one department or the other would already want to own to not have to resort to the project?

a chat that is sometimes so bad to give me a more precise idea about the major criticalities:biggrin:

thanks to all and good work

Edoardo

 

[Tequila]

hi edoardo, these issues depend on multiple factors, the type of new product you want to do, from how the technical department is related with the production/office purchases, and also it is not a standard for all companies.

trying to be a little more precise you should see what kind of product it is, if before the actual production is created a prototype or if as happens on 90% of the companies it sends in direct production and then you will see the to be done.. .

In my case, apart from the processing cycles that I do not manage them ( fortunately), when designing something, I try to use as much as possible components and commercial raw materials, and these materials I place them in the cutting parts. I need this because when I put the distinction in management and a machine is launched in production, we already have the commitments of raw materials and the launch times in production decrease.

the choice of the materials from which to start is the prerogative of the technical department, this to avoid having to build a particular that in order to complete it must be, for example, milled at every point of its own.
I happened to see some plates with measures type 57x37,5 l=700 worked with holes of passage, where the size 57x37,5 did not serve anything and there were no problems of encumbrance, so you could leave quietly from a 60x40 thread, do only the holes and just.

in my specific "loop back" I have them when commercially or for reasons of disproportionate delivery times, some components fail to return them, but fortunately they are almost zero.

for the departments, one thing that asks me is a reference code of a similar component built previously to reduce the time of creation cnc program, or for the revisions of complex carpenters to highlight (with highlighter on paper and with different color on cad) quotas and notes, to intervene only on the points indicated and not have to recheck from scratch a cnc program.
purchases to complete the description of commercial components.

for the leaders to hurry and all right!!!

 

[PierArg]

hi edoardo, the theme you proposed is very interesting.

I feel like listing tequila in full.

If you have specific cases to analyze it would be better because each company is structured to itself. mainly in Italy where small/medium enterprises constitute 90% of the productive fabric.

However, to answer directly to your question about the loop-back I can tell you that the main causes are:
- changes made by the current customer
- a-sample validation test went to "flyfalle"
- modification of materials/processes because the purchasing office imposes it (it is the most widespread point)

 

[yudansha]

hi edoardo, the theme you proposed is very interesting.

I feel like listing tequila in full.

If you have specific cases to analyze it would be better because each company is structured to itself. mainly in Italy where small/medium enterprises constitute 90% of the productive fabric.

However, to answer directly to your question about the loop-back I can tell you that the main causes are:
- changes made by the current customer
- a-sample validation test went to "flyfalle"
- modification of materials/processes because the purchasing office imposes it (it is the most widespread point)

as you underline the "most widespread" point is that of material/process modification.
For example, I was talking to a friend of mine who told me that in his company they had experimented with software that "recommended" where they could optimize the object designed; if not that such software did not take into account that different materials also correspond different technological processes.
from which even if initially a saving could be seen, in fact the production process used certain choices of materials because in the valley too expensive.

also the problem highlighted by tequila regarding the unavailability of materials/components by suppliers is a problem that has attracted my attention.

in general I am trying to investigate (graduate degrees) if there is the possibility to automate/direct certain choices (projects) in relation to choices made by departments "downstream".
that is to move (through cooperation between departments) certain considerations already in the design phase in order to avoid "loop-back" expensive in terms of time (and break :biggrin: )

I therefore ask for help in identifying possible criticalities in this sense (effetically found) that perhaps could be resolved with a greater (or more effective) circulation of information between departments.

Thank you!
I'm sorry.

 

[PierArg]

...in general I am trying to investigate (graduate degrees) if there is the possibility to automate/direct certain choices (projects) in relation to choices made by departments "downstream".
that is to move (through cooperation between departments) certain considerations already in the design phase in order to avoid "loop-back" expensive in terms of time (and break :biggrin: )

...

technical considerations?

Consider that a designer is always at the centre of communication channels between:
- customer (direct or commercial)
- suppliers (direct and/or through purchasing office)
- production/ process engineering: As the project takes shape it is the care of the designer to inform, through meetings etc., the production department to evaluate all possible repercussions (in terms of processes, processing, construction of special equipment for testing or assembly,...) of each individual choice.

As the designer goes on, he informs every institution about what he's doing. when they come back all positive feed-backs will freeze what just done and proceed.

at the end you do the "final review" that "valid" the whole work.

from there is like the classic phrase "someone speaks now or forever silent"

 

[yudansha]

Consider that a designer is always at the centre of communication channels between:
- customer (direct or commercial)
- suppliers (direct and/or through purchasing office)
- production/ process engineering: as the project takes shape is ......

at the end you do the "final review" that "valid" the whole work.

That's what I mean!

so, I would like to understand if there is the possibility to implement an automatic step that replaces "look (production rather than office purchases, etc.ec.) I made this choice, what do you say? Do you approve? "

clearly it would be unthinkable to automate any choice; be sure that each of your department will take considerations to the project because you will arrive at shared solutions.

I wonder whether or not there is something that can be automated.
I know, I throw it there as a superficial example:

"I choose such type of material (project level) that allows me to have smaller thicknesses so less cost during purchase.
the production however goes that to realize this piece is not so convenient because to work it the machine puts us a greater time. "

Perhaps therefore an estimate of working times could be implemented automatically given material and piece size.

I repeat the example is so simple, and maybe you think of something that would benefit in practice.

I hope I have given an idea, otherwise I am here to continue this discussion that I am passionate about:biggrin:

thanks to all and always good work!

I'm sorry.

 

[Fulvio Romano]

in general I am trying to investigate (graduate degrees) if there is the possibility to automate/direct certain choices (projects) in relation to choices made by departments "downstream".
that is to move (through cooperation between departments) certain considerations already in the design phase in order to avoid "loop-back" expensive in terms of time (and break :biggrin: )

you are talking about optimizing business processes. tequila and pierarg, I think I understand, they answered you with operational practices of workshop.

to make a comparison, it is as if I asked the mechanic at the corner to suggest how to optimize the matrix of concordance for the launch of the production of a new car.

Now I don't have much time, maybe later I try to figure out the problem. Maybe start to figure out how a cycle di deming (which you will certainly know already), because you start from the bases, not with a delphi test on a forum :tongue:

 

[PierArg]

I think it's not about the pdca.

I understand that you do not want to automate communication channels but just the information that passes through the channels.

 

[yudansha]

Of course I know the deming circle well

the answers of tequila and pierarg I liked and were useful.

I asked them (and others who intervene) if they recognize that some choices that the designer could take them independently if only he had some data at his disposal.
and therefore ultimately what data?

maybe some of them (data) could automatically arise if only you had a routine that could calculate them (at least in approximation); without passing through an in-depth analysis of the deputy department to provide a "sentence" about it.

 

[PierArg]

That's what I mean!

so, I would like to understand if there is the possibility to implement an automatic step that replaces "look (production rather than office purchases, etc.ec.) I made this choice, what do you say? Do you approve? "

clearly it would be unthinkable to automate any choice; be sure that each of your department will take considerations to the project because you will arrive at shared solutions.

I wonder whether or not there is something that can be automated. . . .

In my opinion you would like to do something that is a hybrid between a management sw (type sap) and a pdm.

where for example management is able to manage technical data and treat them for logistic/productive purposes

 

[yudansha]

si pier, your definition may please me; even if I don't believe much in hybrids. In fact, in this (new) way the development methodologies would be re-arranged.
in any case it is obvious that some things it is possible (perhaps) to implement them already in "superior" levels and others just... And the various departments that are going to make us ehehe!
I would concentrate on things that can be done
and therefore in my opinion it is a good approach to start with those that perhaps are perceived difficulties (from the designer) in terms of "loop-back".
I clearly have no truth and every advice is more than well accepted

I'm sorry.

 

[PierArg]

you need to focus on a tool that can be shared between all departments concerned and where everyone goes to put in the information characterizing that department.
also a simple excel sheet with the appropriate macros is great.

For example, the designer (or who initiates the loop phase) connects a document covering a component/product with its design and code...and of course the request to be placed to other entities.

you have to think of something like a container where everyone, in the appropriate area, goes to put the information.
Once saved, you automatically generate a process (e.g. outlook sending messages automatically) that alerts recipients in real time.

to any occasion/problem/request that you want to address you generate a new container

 

[yudansha]

Thanks pier,
we are on the same wavelength

 

[PierArg]

No. ..even if not I have understood whether you really want to make this tool or you just need to analyze it

 

[yudansha]

I don't think he hesitates. could approach the idea of "concurrent engineering".
can be more or less ambitious depending on what you want to push (and go crazy). Let's say that if there are any issues that can be helped to streamline (that's why I repeat the speech of "loop back") you can try to analyze the problem.
as I told you earlier I am well aware that on some problems you cannot intervene because too rich in variables; but if you find some "loop back" that you can "help", then perhaps you can put to reason on how to set the resolution.

 

[Fulvio Romano]

I asked them (and others who intervene) if they recognize that some choices that the designer could take them independently if only he had some data at his disposal.
and therefore ultimately what data?

For example, the designer (or who initiates the loop phase) connects a document covering a component/product with its design and code...and of course the request to be placed to other entities.

loudly dissented. This would only work in a perfect world. In fact, corporate sharing tools do not exploit these channels.

between the designer that "those heads of c@77o food are not able to read the drawings", and the operator that "those professors who believe that the whole world are in the formulas of their pcs" every single information must be taken with the benefit of the inventory.

It will be that I work in an infamous world, but if there is a problem in an industrial process and I feel twenty people, of which ten operators, five managers and five quality employees, will give me twenty different versions, giving the blame to people and factors that in other versions are instead those that make the difference to quality. truth is not simply an average of all information.

 

[yudansha]

... the truth is not simply an average of all information.

You're right!
but as a decision will be taken, I would think it is a multi objective optimization.
Let me know this simplification, on the other hand we are talking about tools that support not certain of artificial intelligence.

 

[Er Presidente]

a "good friend" advised me to give an eye to this interesting discussion I had "bucato".
I can only express my satisfaction in addressing a topic that "solletic" me the deepest "technological" feelings.
I have to read all previous posts well before I intervene in a complete way.
What I can say now is that the "motivations" for which a project should be subjected to revision, modification or variation can be different and even just trying to list them could be of help.
a first division is the one that divides between "external" and "internal" motivations.
outside are the reasons that arise from the market, suppliers or legislative bodies.
internal are the motivations of corporate efficiency that collects a whole series of aspects.
paying attention to the fact that the various entities within a company tend to behave like a company within the company, so with their microcosm with its internal/outside it is easy to understand that the cycle at the end is strongly iterative and should be explored the hypothesis that the evaluation cycle could also be simple but that the overall complexity of the whole process is not at all.
It's late today, and I have to go to sleep. I'll try to develop the reasoning.

p.s.: unfortunately I have the impression that you will have to prepare for a series of "pistolotti".

 

[MBT]

the drama of all speech is that, in the end, "truths" are extremely aleatorial, as well as concealed... .
Do we make an example?
design a base
I, designer, hypothesized in my head to do it in carpentry. average cost, suitable for small batches, without problems of molds... I realize by bending sheet and performing few welding
the office purchases, or office times and methods if it exists, begins to insult me. wants to give this job to a company that has few bending machines but excellent welding machines! how to wrong him? But who knew?
the carpenter, takes in hand the design and begins to unmadonna. He doesn't find the odds he needs and he has to deal with. But the odds are there! are the ones that I put, besides to build the piece also to verify it, so that "bate" with the other pieces.
similar speech then for the miller.... that you will ask why that wholeness must be +/- 0.05 when its door cutter is ready to reach +/ 0.1 on that measure.... right objection! but I could know? if they gave it to another supplier, the 0.05 could also be done using a cro-magnon operator.. . .
then the assembly arrives... That's funny. "Who is that idiot who mounts the bearings like that, who am I going to get the heater? I beat them with the hammer, which is good the same...." "How do you choose tolerances? with the numbers of the tombola?" and away.... but if they give them to be mounted to another workshop... No problem! evidently the drawings of the details must be "customized" according to the workshop to which they are assigned....

If we get to the test, or to the final user, it's up to Dr. House! Just don't know what...
There, believe me, everything feels and the opposite of everything....
"It worked until yesterday." "No, it never worked." "Yes, it worked, but sometimes it and sometimes it didn't." "It worked very well, then it stopped and you don't know why." "I saw it work, then I went to the bathroom and when I came back it didn't go anymore. "

the designer receives these feedback from the field and his main work is not correcting drawings to adapt them to feedback. his main work is to clear the information, understand where lies, lies, opinability are and, if really necessary, correct the shooting.

Would you like to do this to a software?
If you make it, you need a neural network that matrix is a video game in comparison.... and I run you to the universal nobel and I personally sculpt you an equestrian monument, whose horse will have 6 legs!

 

[Fulvio Romano]

the designer receives these feedback from the field and his main work is not correcting drawings to adapt them to feedback. his main work is to clear the information, understand where lies, lies, opinability are and, if really necessary, correct the shooting.

I wanted to say exactly that at post #16.
and I don't tell you when the manager arrives on the other side of the side of the pavement, that "you're a jerk, you have to do in this other way because I have experience."