Freelance as a means of earning. Part 3 A bit of theory
To begin with - the beginning.
Part 1
Part 2
With the official part done, go to the case.
')
Immediately I will warn you - today I will talk about theboring necessary theory, in the next post we turn to the consideration of the methodology based on the draft - an example.
In the last post I asked in the comments to help me. The fact is that I can’t physically know everything, I can’t foresee all situations and even more so face them. I sincerely regret that no one in the comments began to answer the question I posed - what other markers can be used to define the project stage as a criterion to consider the possibility of starting work on the following. And this is bad, because in fact it turns out a monologue. I would like to engage in dialogue with those who read this series of posts, since it is in the dialogue that one can gain additional knowledge by passing them on to the public. Why this happens - I can not judge, the reasons arise in my head a lot, both simple and offensive. I would like to know - does it make sense in articles to ask some questions to you, readers, or is it a waste of time? (once again he took and asked the question :)))))
Or TOC. Founder - Eliyahu Goldratt. The Israeli physicist is a teacher who at one point looked and said, “Hey! Yes, in business there are the same processes as in physics, only no one pays attention to them! ” At the moment, unfortunately, is dead.
The basis of his theory expires from the data analysis method. That's what we do when we fill up another project? “Unaccounted factors, what can I do?” - you can do something, otherwise you would not read it all. "Well, it happened, next time I will be smarter" - is it true? Or, in general, the epic - “So the stars were formed”. By the way. I hope none of you are not against the fact that he himself is to blame for the failure of the projects, and not some external factors? This I chewed in the first parts, for which way, and received several reproaches. If you do not agree - either re-read the beginning, or do not read at all - meaning?
Something I digress. In general, we are not inclined to analyze the problem from the point of view of banal logic. In physics, having received a completely different result than expected, scientists do not omit the result as an unsuccessful example (“Yes, this is Zeus with a lightning heracnul”), but seek him a simple and understandable explanation. Well, or not quite simple, otherwise we would all be experts in quantum physics, but nonetheless fit into certain frameworks and laws.
And this very uncle, terribly intelligent from physical knowledge, and with an appropriate way of thinking, decided to take up the business. Yes, he took it so that a huge number of firms used his advice, his literature became the reference book of devotees in the theory of restrictions, and his weight and authority were not recognized and recognized in a single sphere.
The essence of his theory is that in any production process there is a certain limiting factor, the weakest link, which sets the pace of work of the entire chain. Assume that at the factory the part passes successively through the machines 1, 2 and 3. Machine 1 makes 100 parts per hour, machine 3 - 300 parts per hour, machine 2 - 10. Accordingly, the performance of the entire line is 10 parts per hour.
"Yes, it's banal," you exclaim. Duc. What and speech. Just to say that everything is trivial, and taking action to remove this banalism is two different things. As far as actions are concerned, everything is much more complicated here, as it has been refined psychologically for a long time.
So, we have determined that machine 2 in our country is a restriction of the chain, setting the rhythm of the entire line. The first thing that comes to mind is to increase its throughput by simply increasing the number of machine tools of type 2. Now you ask - yes, what, nafik, machines ?! We have programming! Patience, we'll get to it, just an example of machine tools is simpler in perception (for it is visual), as well as TOC requires a revision in the root of the worldview, then - from simple to complex. In the meantime, just try during the reading to try to understand how this can be applied to the practice of software development.
An increase in the number of machines is a solution, of course, but not the fact that there are funds for it, nor is it a fact that the overhead will not increase.
The second method is not so transparent, and unpleasant at the level of physiology, but nevertheless it is so - it is necessary to limit production on machine 1. In its current form, we only create mountains of reserve at machine 2, while tying up capital. Why is it difficult? Because if a person does not work all the time, then he is useless, and we immediately begin to look for ways to reduce it. The kind of body hanging around leads us into a subconscious panic, since we most likely did not take into account everything that needs to be done, which means that the salary is paid just like that, and it is quite possible to do any work at the expense of a person who is bored.
In fact, it is not. The task is to squeeze everything that is possible from machine 2, while expecting that it will be a restriction for the chain, and the productivity must be adjusted to its needs. In addition to limiting production (why do more than we can actually recycle?), It is necessary to provide a reserve in case of a breakdown of the machine 1 so that work does not stop in any way. This stock is called a buffer.
In addition, it is possible to increase the productivity of the machine 2 by placing a quality control in front of it so that it does not process parts that are obviously defective. The task is to understand that the operating time of the machine 2 is the most valuable, and it is reasonable to spend every minute of its work. Even to the detriment of the imaginary rationality of the work of other machines.
We looked at one line with you. Now let's imagine a more complex production. For example, we release a certain device. Although ... Why some. We release TV remotes. we have line 1, which produces cases with a capacity of 1000 pieces per hour. We have a line that produces stuffing in the form of soldered boards with a capacity of 100 pieces per hour. And we have a line of rubber products, producing buttons with a capacity of 1000 pieces per hour. After that, all the details fall on the assembly and we get the coveted remote from the TV - the object of desire, if judged by modern folklore, of every man.
The first thing to understand is the performance (current) of our production - 100 pieces per hour. fasten lines 1 and 3 to release 100 products per hour.
Second, let's take a closer look at line 2. We have a process of cutting blanks, which gives out 500 blanks of boards per hour, there is a machine for etching / machining a printed circuit board and screen printing with a capacity of 100 boards per hour, and there is a machine for desoldering, which gives 300 boards at one o'clock. What will we have if everything works at 100%? Machine 1 will operate at full capacity, devouring textolite reserves and delivering the associated reserves at a tremendous speed (reserves that are accumulated in the form of cut blanks — they cannot be put anywhere, they all wait for processing and nothing more). Machine 2 will puff on its 100%, and machine 3 will constantly complain about the lack of components. Zvizdets.
The first thing to do is to determine the percentage of barracks received on the machine 1. Let the uncle wield an old machine, giving one damn detail to a dozen valid ones. So, we need to fasten the power of the machine 1 to 110 per hour (it is better to vary it a bit to give the opportunity to provide a day supply in case machine 1 stands idle.
Next - you need to put a line of quality control before the machine 2, which will identify this defect. Why make a fee for parts that are already automatically a marriage?
Then it is necessary to coordinate the actions of the machine 3 in such a way (to make a schedule, etc) so that it is loaded 100% of the output from the machine 2. Thus, we squeeze everything that we have from line 2.
Is it logical Yes of course. Everything is very transparent and understandable, some kind of banality is written. Well, if you think so, then let's see what we do.
Consider the development of the site. Work on the project - designer, layout designer and programmer. The specificity of the site is such that we are obliged to download the maker-up to the ears. Think of the reason yourself - a bunch of drop-down menus, specific tables - but you never know what. I am sure that many will be able to give an example of such a project. Separately, I will touch on the terms and costs given in the draft — all this is a figment of my imagination, and is not at all a subject for discussion.
While we are developing the concept of design, everything is going fine - the start of work, and everything seems to be according to plan. After this, the development of the site pages begins. The designer performs one village, say, for 1 day. The coder needs 3 days to issue the html page. The programmer spends 1 day to revive this page and fill it with data. The designer page costs 2 tr, the page of the layout maker - 4 tr, the page of the programmer - 2 tr. The total number of pages on the project - 30.
I think from the statement of the question it will become clear to what I am leading now. And you are absolutely right - when the designer starts working on the pages without interruption - we have become similar to line 2 from the previous example.
I will explain a little. As a rule, the designer performs the work until they run out. those. from the beginning of the work on the pages to their end will be 30 days. In this case, the coder can actually handle only 10 pages during this time. Those. you have frozen assets 60 tr ... Congratulations.
Why? Because you paid off with the designer for the work, the implementation of which is optional. The layout time will be 90 days, of which for 60 days the layout designer will dig debris from the graphics that have accumulated in front of him. Those. it is a constraint in the whole chain of operations. The size of the buffer in 30 days, when the designer will hand over the work, will be maximum (the buffer will have chopped graphics for the 60 days of work of the layout maker and 60 tr.).
Now imagine that a person comes to you with work for 10 days. To do the work you need to spend 20 tr. Which, given the fact that 60 tr. frozen in the form of graphics, which is now really not needed, you do not. We have lost the client.
In addition, imagine (again, our example is nothing more than fiction, so that you can imagine anything) that the designer can work with graphics. Because of ignorance, or some other - it does not matter. But it is quite possible that some of the pictures transmitted to them may not be suitable. What happens in this case? The maker-up, faced with a problem, will wrap up the schedule, requiring corrections. A designer, who has not thought about the project for a month, suddenly finds out that he sent the wrong thing. The formatter will quit the job done halfway, and proceed to the next stage. The programmer will say that "you are so bold, I work three days later, and here you are not in the order you agreed on." While you receive a package from the designer - time will pass. Yes, time is compensated by the fact that we have created an enormous buffer size in front of the coder, if this is, of course, not the completion of work. See the problems and nerves that will fall on you? Doesn't it look like an uncle with his machine tool for cutting parts and solving a problem?
Actions that seemed so obvious to us suddenly lose that very obviousness. Not? Tell me, will it be a normal practice for you to agree with the designer that he sends the materials in packages, each of which is paid separately? And he does this for 3 months, and not one as he expected? And will it be easy for you to introduce the acceptance of materials from the designer without distracting from the work of the layout designer?
Naturally, you can say that my example is invented and too simplified and exaggerated. Naturally. If I start from the very beginning to push the methodology of the critical chain without preliminary preparation, then, most likely, I will receive a bunch of negative reviews. Since, in fact, even though the real project is different (not so simplified and exaggerated), all the same, the ideas voiced here will be reflected in it.
So I suggest you think about how well the situation I described and the measures taken are justified (even though you agreed that it works in the case of machine tools) and expect the next publication. As I promised, it will analyze the project close to the real one, and build the methodology for its implementation.
Thank.
Part 1
Part 2
With the official part done, go to the case.
')
Immediately I will warn you - today I will talk about the
A little whining
In the last post I asked in the comments to help me. The fact is that I can’t physically know everything, I can’t foresee all situations and even more so face them. I sincerely regret that no one in the comments began to answer the question I posed - what other markers can be used to define the project stage as a criterion to consider the possibility of starting work on the following. And this is bad, because in fact it turns out a monologue. I would like to engage in dialogue with those who read this series of posts, since it is in the dialogue that one can gain additional knowledge by passing them on to the public. Why this happens - I can not judge, the reasons arise in my head a lot, both simple and offensive. I would like to know - does it make sense in articles to ask some questions to you, readers, or is it a waste of time? (once again he took and asked the question :)))))
Theory of limitations
Or TOC. Founder - Eliyahu Goldratt. The Israeli physicist is a teacher who at one point looked and said, “Hey! Yes, in business there are the same processes as in physics, only no one pays attention to them! ” At the moment, unfortunately, is dead.
The basis of his theory expires from the data analysis method. That's what we do when we fill up another project? “Unaccounted factors, what can I do?” - you can do something, otherwise you would not read it all. "Well, it happened, next time I will be smarter" - is it true? Or, in general, the epic - “So the stars were formed”. By the way. I hope none of you are not against the fact that he himself is to blame for the failure of the projects, and not some external factors? This I chewed in the first parts, for which way, and received several reproaches. If you do not agree - either re-read the beginning, or do not read at all - meaning?
Something I digress. In general, we are not inclined to analyze the problem from the point of view of banal logic. In physics, having received a completely different result than expected, scientists do not omit the result as an unsuccessful example (“Yes, this is Zeus with a lightning heracnul”), but seek him a simple and understandable explanation. Well, or not quite simple, otherwise we would all be experts in quantum physics, but nonetheless fit into certain frameworks and laws.
And this very uncle, terribly intelligent from physical knowledge, and with an appropriate way of thinking, decided to take up the business. Yes, he took it so that a huge number of firms used his advice, his literature became the reference book of devotees in the theory of restrictions, and his weight and authority were not recognized and recognized in a single sphere.
The essence of his theory is that in any production process there is a certain limiting factor, the weakest link, which sets the pace of work of the entire chain. Assume that at the factory the part passes successively through the machines 1, 2 and 3. Machine 1 makes 100 parts per hour, machine 3 - 300 parts per hour, machine 2 - 10. Accordingly, the performance of the entire line is 10 parts per hour.
"Yes, it's banal," you exclaim. Duc. What and speech. Just to say that everything is trivial, and taking action to remove this banalism is two different things. As far as actions are concerned, everything is much more complicated here, as it has been refined psychologically for a long time.
How to deal with restriction
So, we have determined that machine 2 in our country is a restriction of the chain, setting the rhythm of the entire line. The first thing that comes to mind is to increase its throughput by simply increasing the number of machine tools of type 2. Now you ask - yes, what, nafik, machines ?! We have programming! Patience, we'll get to it, just an example of machine tools is simpler in perception (for it is visual), as well as TOC requires a revision in the root of the worldview, then - from simple to complex. In the meantime, just try during the reading to try to understand how this can be applied to the practice of software development.
An increase in the number of machines is a solution, of course, but not the fact that there are funds for it, nor is it a fact that the overhead will not increase.
The second method is not so transparent, and unpleasant at the level of physiology, but nevertheless it is so - it is necessary to limit production on machine 1. In its current form, we only create mountains of reserve at machine 2, while tying up capital. Why is it difficult? Because if a person does not work all the time, then he is useless, and we immediately begin to look for ways to reduce it. The kind of body hanging around leads us into a subconscious panic, since we most likely did not take into account everything that needs to be done, which means that the salary is paid just like that, and it is quite possible to do any work at the expense of a person who is bored.
In fact, it is not. The task is to squeeze everything that is possible from machine 2, while expecting that it will be a restriction for the chain, and the productivity must be adjusted to its needs. In addition to limiting production (why do more than we can actually recycle?), It is necessary to provide a reserve in case of a breakdown of the machine 1 so that work does not stop in any way. This stock is called a buffer.
In addition, it is possible to increase the productivity of the machine 2 by placing a quality control in front of it so that it does not process parts that are obviously defective. The task is to understand that the operating time of the machine 2 is the most valuable, and it is reasonable to spend every minute of its work. Even to the detriment of the imaginary rationality of the work of other machines.
From simple to complex
We looked at one line with you. Now let's imagine a more complex production. For example, we release a certain device. Although ... Why some. We release TV remotes. we have line 1, which produces cases with a capacity of 1000 pieces per hour. We have a line that produces stuffing in the form of soldered boards with a capacity of 100 pieces per hour. And we have a line of rubber products, producing buttons with a capacity of 1000 pieces per hour. After that, all the details fall on the assembly and we get the coveted remote from the TV - the object of desire, if judged by modern folklore, of every man.
The first thing to understand is the performance (current) of our production - 100 pieces per hour. fasten lines 1 and 3 to release 100 products per hour.
Second, let's take a closer look at line 2. We have a process of cutting blanks, which gives out 500 blanks of boards per hour, there is a machine for etching / machining a printed circuit board and screen printing with a capacity of 100 boards per hour, and there is a machine for desoldering, which gives 300 boards at one o'clock. What will we have if everything works at 100%? Machine 1 will operate at full capacity, devouring textolite reserves and delivering the associated reserves at a tremendous speed (reserves that are accumulated in the form of cut blanks — they cannot be put anywhere, they all wait for processing and nothing more). Machine 2 will puff on its 100%, and machine 3 will constantly complain about the lack of components. Zvizdets.
The first thing to do is to determine the percentage of barracks received on the machine 1. Let the uncle wield an old machine, giving one damn detail to a dozen valid ones. So, we need to fasten the power of the machine 1 to 110 per hour (it is better to vary it a bit to give the opportunity to provide a day supply in case machine 1 stands idle.
Next - you need to put a line of quality control before the machine 2, which will identify this defect. Why make a fee for parts that are already automatically a marriage?
Then it is necessary to coordinate the actions of the machine 3 in such a way (to make a schedule, etc) so that it is loaded 100% of the output from the machine 2. Thus, we squeeze everything that we have from line 2.
Is it logical Yes of course. Everything is very transparent and understandable, some kind of banality is written. Well, if you think so, then let's see what we do.
We are horrified together.
Consider the development of the site. Work on the project - designer, layout designer and programmer. The specificity of the site is such that we are obliged to download the maker-up to the ears. Think of the reason yourself - a bunch of drop-down menus, specific tables - but you never know what. I am sure that many will be able to give an example of such a project. Separately, I will touch on the terms and costs given in the draft — all this is a figment of my imagination, and is not at all a subject for discussion.
While we are developing the concept of design, everything is going fine - the start of work, and everything seems to be according to plan. After this, the development of the site pages begins. The designer performs one village, say, for 1 day. The coder needs 3 days to issue the html page. The programmer spends 1 day to revive this page and fill it with data. The designer page costs 2 tr, the page of the layout maker - 4 tr, the page of the programmer - 2 tr. The total number of pages on the project - 30.
I think from the statement of the question it will become clear to what I am leading now. And you are absolutely right - when the designer starts working on the pages without interruption - we have become similar to line 2 from the previous example.
I will explain a little. As a rule, the designer performs the work until they run out. those. from the beginning of the work on the pages to their end will be 30 days. In this case, the coder can actually handle only 10 pages during this time. Those. you have frozen assets 60 tr ... Congratulations.
Why? Because you paid off with the designer for the work, the implementation of which is optional. The layout time will be 90 days, of which for 60 days the layout designer will dig debris from the graphics that have accumulated in front of him. Those. it is a constraint in the whole chain of operations. The size of the buffer in 30 days, when the designer will hand over the work, will be maximum (the buffer will have chopped graphics for the 60 days of work of the layout maker and 60 tr.).
Now imagine that a person comes to you with work for 10 days. To do the work you need to spend 20 tr. Which, given the fact that 60 tr. frozen in the form of graphics, which is now really not needed, you do not. We have lost the client.
In addition, imagine (again, our example is nothing more than fiction, so that you can imagine anything) that the designer can work with graphics. Because of ignorance, or some other - it does not matter. But it is quite possible that some of the pictures transmitted to them may not be suitable. What happens in this case? The maker-up, faced with a problem, will wrap up the schedule, requiring corrections. A designer, who has not thought about the project for a month, suddenly finds out that he sent the wrong thing. The formatter will quit the job done halfway, and proceed to the next stage. The programmer will say that "you are so bold, I work three days later, and here you are not in the order you agreed on." While you receive a package from the designer - time will pass. Yes, time is compensated by the fact that we have created an enormous buffer size in front of the coder, if this is, of course, not the completion of work. See the problems and nerves that will fall on you? Doesn't it look like an uncle with his machine tool for cutting parts and solving a problem?
Actions that seemed so obvious to us suddenly lose that very obviousness. Not? Tell me, will it be a normal practice for you to agree with the designer that he sends the materials in packages, each of which is paid separately? And he does this for 3 months, and not one as he expected? And will it be easy for you to introduce the acceptance of materials from the designer without distracting from the work of the layout designer?
Naturally, you can say that my example is invented and too simplified and exaggerated. Naturally. If I start from the very beginning to push the methodology of the critical chain without preliminary preparation, then, most likely, I will receive a bunch of negative reviews. Since, in fact, even though the real project is different (not so simplified and exaggerated), all the same, the ideas voiced here will be reflected in it.
So I suggest you think about how well the situation I described and the measures taken are justified (even though you agreed that it works in the case of machine tools) and expect the next publication. As I promised, it will analyze the project close to the real one, and build the methodology for its implementation.
Thank.
Source: https://habr.com/ru/post/149360/
All Articles