Games

Why do we need models and simulations. What is a business model and why does your business need it. Of course, the same model can be included in different classes, depending on the attribute on which the class and fication are maintained.

It is a very common way of analyzing and forecasting the economic situation. Moreover, economic models can be applied both at the level of an ordinary entrepreneur or investor, and at the level of large companies, states, and when studying the processes taking place in the global economy.

The essence of economic modeling is to build a simplified scheme of processes occurring in a certain area of the economy and highlight the most important factors in a compact and concise form.

Building an economic model requires compliance with a number of factors, these include:

— realistic assumptions made

- the ability to predict

— sufficient information support

- Possibility of practical testing.

In different cases, different complexes of these requirements are priority, it is quite difficult to build a model that fully corresponds to all of them, and the need for this arises quite rarely. This is due to the fact that the main goal of economic modeling is the practical application of models and, depending on the requirements, the priority requirements for the properties of the model also change.

Process building an economic model goes through a series of stages. There are three main stages:

Selection of used variables
Making the necessary assumptions
Identification of the main hypotheses that explain the relationship between model parameters.

Variables - specific data, which form the basis of the model, they are divided into exogenous and endogenous. That is, internal and external. Assumptions make it possible to simplify a number of processes occurring in the model and thus simplify the model itself and speed up the process of its creation.

Nowadays, two types of economic models are most common - equilibrium and optimized. Optimized ones are mainly used in marketing research, market research. In such models, various marginal indicators most often appear, such as marginal income, marginal utility. Often this modeling method is called marginal analysis.

Equilibrium models are used to study the relationship between various objects of the economy. The main assumption in such models is that any modeled system is in equilibrium and factors that can bring it out of equilibrium are not taken into account. Typically, the construction of economic models of this type is used to study different markets and the interaction of companies operating in the same market.

It is equilibrium models that are most applicable to private entrepreneurs and investors, since with their help they can obtain valuable information about the market in which they operate and the prospects for its development.

In addition to these varieties of models, they are also divided into positive and normative. In positive models, the main purpose of construction is to find the causes and consequences of an event or economic phenomenon. At the same time, these phenomena are not evaluated.

Normative models, on the contrary, allow assessing a phenomenon or event, but do not allow establishing the causes and consequences of this phenomenon. Both types of modeling are interconnected and are used simultaneously for the most accurate modeling of economic processes.

Do you use economic models in your activities?

Andrey Malakhov, professional investor, financial consultant

As mentioned above, there are many reasons why political scientists resort to the use of mathematical models. However, this method has both disadvantages and advantages. Modeling is a process of simplification and deductive inference. Simplification entails the loss of information about the event. Deductive inference often involves complex mathematical processing, which, at least at first, makes it difficult to work with the model. Therefore, in relation to modeling, a reasonable question arises: why do we need all these difficulties?

The first reason that prompts us to model political behavior is that the model helps to formalize the events taking place in society. The point is that political life is sufficiently regular for a simplified informal model to be of some use. Most of what happens in the realm of politics is usually not is completely unexpected - in fact, the presence of an element of surprise indicates that we have a priori ideas about how events can develop, and we are able to recognize the fact of an unexpected turn of events. So, in our brains we have a kind of mental models of the functioning of political systems, even if we have never tried to express them explicitly. Mathematical models just help to explicate such informal models.

An example of a mental model is the following. Suppose that in the upcoming presidential election one of the candidates is gaining 95% of all votes. Obviously, this does not contradict either the constitution or the established electoral procedures. However, we will be inclined to consider such a fact as extremely unlikely for a number of reasons. First, we assume that there will be a sufficient number of voters from each party to minimize the possibility of a purely random voting result. Secondly, we proceed from the fact that no party will nominate such an unpopular candidate that he could collect only 5% of the votes. Thirdly, we believe that the votes are counted without fraud. One could list more, but the bottom line is that we have a number of assumptions about the US political system, in the light of which the division of votes into 5 and 95% seems to us implausible.

All such assumptions simplify reality. We do not know what the exact number of voters is, and we do not need it - we just know that it is very large. We do not know what specific features of a candidate make him acceptable to some voters and unacceptable to others, but we assume that completely unpopular candidates will not be put to the vote. Few people have enough personal experience in counting votes to know if elections are fair, but all past experience suggests that there is no place for fraud in elections. 2 . Since these assumptions don't often lead us to the wrong conclusions, we can use this model political system for informal forecasting of the future. In fact, when a candidate receives 95% of the votes, there is strong distrust among the population, sometimes even to the point of demanding an investigation, so our model also partially determines the actions and attitudes of people.

Another reason for using mathematical modeling is the need to explicitly describe the mechanisms that explain our informal predictions. Despite the fact that all individuals know what can and cannot be expected from a given political system, they are often unable to determine exactly why And what exactly they expect from her. The formal model just helps to overcome the overly loose formulations of the assumptions of the informal model and give an accurate, and sometimes verifiable forecast.

The above example is derived from the Downs model, which we will discuss later in this chapter. The Downs Formal Model predicts that any political party in an alternative election will choose its candidates and platform in such a way as to attract the largest possible number of voters with their help. This and some additional considerations lead us to the conclusion that there is a tendency for political parties to receive approximately equal numbers of votes in elections; this is the outcome commonly seen in US elections. Thus, this formal model predicted not only that a 95:5 distribution of votes was unlikely, but also that a 50:50 distribution would be expected, for which some justification was given.

Sometimes, it seems that mathematical models just confirm the already obvious things. In fact, this is an inherent feature of any models insofar as they are expected to reproduce, to one degree or another, everything that happens in everyday political reality. However, people tend to have a very vague idea of what "obvious" is. Consideration of a number of contradictory aphorisms (“a wolf smells a wolf from afar” and “extremes converge”, “with out of sight, out of mind” and “the farther out of sight, the closer to heart”, etc.) convinces us that common sense often turns out to be correct precisely because it is so vague that it simply cannot be wrong .

The rigor of formal models, on the contrary, means precisely that they can be wrong, and as a result, the “sports performance” model can sometimes be worse than more ambiguous common sense. However, this is not a weakness at all, but, on the contrary, an advantage of modeling, because the assumptions and forecasts of the model turn out to be accurate enough to be checked, as well as to indicate where and how a possible error occurred. The model that has withstood a number of attempts to distort it is quite likely to give correct forecasts in the future. A model that gives incorrect predictions over and over again, apparently, should be eliminated from consideration.

In short, a model is only useful if, in principle, it can be shown to be wrong. If it is impossible to show that the model is wrong, then it is also impossible to prove that it is correct, and hence the conclusion that such a model is useless follows. An informal intuitive model that allows you to avoid all sorts of mistakes can be a great tactical help in negotiations, but it is powerless to help us understand the mechanism of political behavior more clearly.

A third advantage of formal models over mere intuition or even well-reasoned natural language reasoning is their ability to deal systematically with entities of a higher level of complexity. Natural languages (like English) originated as means of communication, not as means of inference. Mathematics, on the contrary, was originally conceived as a means of logical inference and the systematic operation of concepts. And experience has shown that mathematics is a very useful tool in this respect. Political scientists, for their part, are only now beginning to realize what modeling can provide for a deeper understanding of political behavior, and in some cases entire branches of mathematics should have developed (the most notable example is game theory) before social scientists could see something in common in disparate types of social behavior. Mathematical modeling of social behavior is no more than 20 years old, and so far there is no reason to believe that it has already reached the limits of its development.

Finally, the advantage of mathematical modeling is also that it allows different scientific disciplines to exchange their research tools and techniques. There are many examples of this: in the models used in political science, not only basic mathematical tools are involved, but also a lot of techniques borrowed from econometrics, sociology and biology. Polling - which is essentially a complex mathematical model of the distribution of public opinion among different groups of the population - is a widely used method used in most social sciences. Borrowing occurs in the opposite direction: system engineers, developing large computer models of global socio-demographic processes, were forced to turn to political science models to clarify political aspects, and more recently, mathematicians working on a new theory of chaotic behavior discovered that the Richardson race model weapons (see example 1) lends itself to a very productive analysis using the methods of the above theory. Similarly, game theory was originally developed by economists and political scientists to analyze the phenomenon of competition and only later turned into a branch of pure mathematics.

In addition to stimulating the interdisciplinary exchange of methods and ideas, mathematical models are also useful in that they allow us to see the deep homogeneity of phenomena that at first glance have nothing in common with each other. The following example, quite trivial in itself, clearly demonstrates this type of generalization.

Imagine a simple game in which two players take turns taking chips from the table, numbered from 1 to 9:

1 2 3 4 5 6 7 8 9

The winner is the one who first collects chips for an amount equal to 15. Playing this game, you will undoubtedly find that it has its own tricks - in particular, in the order of a defensive move, you can take exactly those chips that the second person needs from the table. player to get the final amount - however, the overall strategy of the game, apparently, is not entirely obvious. To generalize the game, we rewrite the chip numbers as follows:

4 3 8

9 5 1

2 7 6

Note that in such an entry, each row, column, and diagonal sums up to the desired outcome - 15. Thus, for a successful game, you need to choose one of these rows of numbers. In this form, the game already looks very familiar: it's tic-tac-toe, which any five-year-old child can play. After we presented the game in a streamlined way, what at first seemed unfamiliar to us now began to look quite recognizable, so that we were able to use a well-known solution for a long time in a new context.

This exercise - of course, in more complex forms and applied to more significant problems - is very characteristic of the process of finding commonalities using mathematical models. Many cases are known when a mathematical model, developed initially for one problem, turned out to be equally applicable to other problems. For example, Richardson's model of the arms race can be used to study not only the international arms race, but also the dynamics of growth in campaign spending by rival political parties, or the process by which bidders inflate the price of a "tidbit" product. The prisoner's dilemma game applies not only to the example of trench warfare (see below), but also to the case of a "price war" between two gas stations, and also to the case of a government decision to develop a new type of weapon. A variation of the prisoner's dilemma game called "chicken" originates from the games of young thugs running in wrecked carts along the abandoned roads of the California desert; she is now applied to the study of the policy of nuclear deterrence under the threat of thermonuclear war. The list of examples could be endless; for us, however, it is significant that most good mathematical models find applications far beyond the problems for which they were originally developed.

So, mathematical models have four potential advantages over natural language models. First, they order the mental models that we usually use. Secondly, they are devoid of inaccuracy and ambiguity. Thirdly, mathematical notation, unlike natural language expressions, allows one to operate at a very high level of deductive complexity. And, finally, mathematical models contribute to finding common solutions to problems that seem heterogeneous at first glance.

Classic situation: The customer gives input data and expects us to evaluate the project. Often sends a document proudly called "Technical task". Sometimes, however, this is really the Terms of Reference. Only in most cases, even an adequate input document in no way guarantees an adequate assessment from a potential contractor.

Why can't the contractor give an accurate assessment of the project if there is a full Terms of Reference?

If every specific firm with a specific CFO or CIO were not unique regardless of the field of activity, and any, except for the already generally accepted, operations were not done as convenient for the authorities / as it happened historically - then yes, TK would be enough. But every time, after reading the documentation, our specialists start asking questions, it turns out the sea of nuances that you can’t fix in a day. And this is if we have a project planned for any one contour. What if there are several?

Can object - give an assessment right away, you are experienced, you have seen everything. All right. Only if the specialist evaluates immediately, he will lay down all conceivable risks, remember all his most problematic clients and projects, and throw another 20 percent on top, so that he is guaranteed not to work for free, and not to fight in the end with the Customer, knocking out additional money from him. financing.

Hence the need for a survey is born, which is rarely free of charge, and sometimes costs 10-15% of the budget of a future project.

But for greedy 1C nicknames, this is not enough! After the survey, at best, they give an estimate of the future project with a delta of 30-50%. Why? Have you found out, studied, what else is needed?

Indeed, at the survey stage, specialists in the interview mode find out what is happening with the Customer, how the work is being done now, what are the wishes for the future system, what they like, what they don’t.

But the result of the survey is the information collected in the report. There are no guarantees that the management needs exactly what the key users want, and even what the management itself verbally told the specialist. Very often, business customers want to see how the work of their departments in the new system will look like. How current operations will be reflected, some of which may still be done manually. And often the vision of business customers is very different from the implementation of such functions in a typical version of the system. And no survey can give an unambiguous answer - do the visions of the future Customer and the Contractor coincide, or after the first demonstration of the modified system, it will have to be redone by 50%, because. “It’s not like that with you, and in general I didn’t want it.” And this risk will certainly be included in the project assessment, again greatly increasing it. However, after the examination, it is usually possible to determine the upper bar after the survey, although the numbers very often frighten Customers, up to the refusal to continue work, despite the fact that this is a theoretical maximum - it is often perceived as a final assessment.

How to be? It is necessary to make sure that the vision of the future system by the Customer and the Contractor coincides before the start of the main work on the project. The customer must see the future system, understand how his processes will be reflected in it, and the project team must find out in advance which processes will “fall” on the standard functionality, and which systems need to be finalized (or vice versa - which processes the Customer is ready to adapt to the standard functionality ).

One of the options for achieving these goals is the stage of modeling - a process-by-process comparison of all business processes of the Customer with their reflection in the system intended for implementation and demonstration of end-to-end examples. As a result, a document is usually formed called “Functional Coverage Map”, containing in tabular form a register of processes and all identified inconsistencies with reality and implementation in the system, as well as a brief description of future improvements, a detailed description of which will be, in fact, the Terms of Reference.

I always insist on doing simulations, because for adequate money, both parties will end up with the same vision of the future project.

In general, this stage is primarily needed for:

Minimizing the risks of the parties

Obtaining the most adequate assessment and the formation of a common vision of the result of the future project among all its participants.

Based on the results of the simulation, the Contractor and the Customer:

They finally understand what will be the difficulties in working with these specific people on the other hand

They get the final score, which is very likely to be less than the one given by the results of the survey.

And although the cost of this stage can sometimes reach up to 30% of the budget of the entire project, the benefit from its implementation primarily for the Customer, as practice has shown, is very significant both financially and functionally.

Recently, other things being equal, I do not sell the examination stage as a separate one - I immediately offer a simulation with the examination included in it. The output forms of this stage (functional coverage map) are much more informative both for the Customer and for any contractor.

It is time to return a little to the cycle of materials that were discussed last summer. This is necessary in order to put an end to that cycle with today's material (and start a new one with peace of mind).

So what was the summer like?

We started the cycle with
Then we looked at the work of this intellectual tool on contextual advertising
After a particular case with contextual advertising, we looked at how you can apply
This allowed us to start (there are limits to the applicability of intelligent tools?)
After we moved on to (any system becomes complicated where there is more than one feedback - that is, wherever a person appears, a complex system immediately arises)
To influence the chaos, (they will allow you to be able to influence this happening better)
And having made such a big circle, we returned again to the use of intelligent tools for solving particular applied problems (already from the point of view)
This allowed us to confidently consider the topic (in order to predict the future of these systems)

At the same time, by an amazing coincidence, we bypassed the question: “What is a model?”.

In a general sense, a model is a kind of description of a process or event. In business, the most famous are business models (a description of how exactly the owner will make money with his business) and business process models (for example, a description of exactly how, when, to whom and why Fatima should offer a pie at the McDonald's checkout).

There can be many models. But in the beginning, simple models will be enough to solve applied problems.

In order not to complicate your life when working with models, it is useful to adhere to the following criteria:

Models should be simplified - they should not cover all aspects of reality, but only the most significant
Models should be pragmatic - that is, focused on what is useful at the moment
Models should generalize—that is, provide a summary of complex relationships
Models should be visual - that is, they should visually explain what is difficult to explain in words (this also increases their usefulness when communicating with colleagues, managers and subordinates)
Models should order - that is, structure information and put it on the shelves
Models should be a working tool - they should not give ready-made answers. No. Their first and foremost task is to ask questions. And only when you start working with this or that model, answers will appear.

What are models for?

When our brain encounters chaos, it automatically (!) starts to create systems in order to recognize this chaos, structure it, or at least get the fullest possible picture of what is happening. That is why people always find explanations for what happened (which leads to the wilds of myths like lightning from the sky, as a sign of the wrath of the gods). That is, it happens independently of us. People just can't help but react. The neocortex is constantly working, building up a picture of the future and constantly trying to predict the future. This is an element of evolution that constantly leads us into blind alleys of inertia of thinking and instrumental blindness.

Models help us make this task easier. Because building models is a conscious process. It forces you to drop the secondary and concentrate on the most important.

Critics like to point out that models do not reflect reality. It's right. But it is wrong to say that models contribute to the standardization of thinking. On the contrary, the model is the result of logical thinking, which requires conscious active effort. And that is why building a new or applying an existing model often helps to go beyond the inertia of thinking. This is the importance of the model.

Two approaches to using models

There are two approaches to use models. The so-called "American method" and "European method".

Americans love to try and make mistakes. The ideal of this approach is Edison. The standard of this approach is to make as many mistakes as possible per unit of time. This training is completely hands-on. Attempt, failure, conclusions, new attempt. This is far from always productive (but in).

Europeans, on the other hand, tend to first get acquainted with the theory, and then do something and fail. After that, they analyze what they have done, correct the mistakes and try again. Here the process is somewhat different. First we read the instructions, then we put it into practice, if we fail, we draw conclusions, study the theory more carefully and put it into practice again. The use of such an approach in solving simple problems is redundant in terms of resources. But it allows you to more gracefully solve complex problems.

Approaches are neither good nor bad. They just are. And it is important to remember the main rule:
Each model is only as good as its performer.

Liked? Share!

Creativity is in every child, the main thing is to be able to discern it, to help its development. Modern children have many opportunities to show their imagination: someone likes to draw, someone does crafts, someone writes stories. And, of course, many are interested in all kinds of equipment: military, automotive, construction, aviation. Is there such a hobby in the world that combines all these types of creativity? Exists! Modeling is such a universal hobby - assembling models of various types of equipment. If you have not yet decided on a hobby for your child, it is quite possible that you should learn more about this interesting activity.

Assembly of models, or modeling - what is it?

Do not confuse this concept with "modeling" (designing clothes, or creating computer images using 3D graphics), or with "simulation" (building and studying models of real-life phenomena and processes - physical, chemical, etc.).

Modeling is a hobby that not only children, but also adults are willing to do, since this activity can be of different types of complexity and is generally quite diverse. For example, there is bench modeling. Simply put, this is an assembly of models of a wide variety of equipment from different materials (paper, wood, plastic, metal) in order to be able to play with them, collect them, or simply admire them by putting them on a shelf. This also includes, for example, the manufacture of tin soldiers.

Another type of modeling is the assembly of models of cars, planes, helicopters, ships that can move and are exact reduced copies of real-life equipment. This more complex type of assembly of models belongs to sports and technical modeling, because in many countries, and more recently in Russia, this hobby has, in fact, turned into a new sport in which international competitions are held.

Assembling models of equipment with children. When to start? Where to start?

At what age is it best to start tinkering with assembling models? Oddly enough, from a very young age. The kid receives the first design skills by playing ordinary cubes, putting together puzzles, assembling children's designers. It is best to start modeling directly by buying prefabricated paper models of houses, castles, airplanes and cars for kids, and then move on to more “serious” materials.

What are the benefits of modeling for children?

First of all, these are, of course, the skills of precise manual labor, accuracy, attentiveness, and perseverance. In addition, assembling models can bring real pleasure, both from the very process of creating interesting crafts, and from the result - your own gallery of various models, ranging from the simplest animals, geometric shapes, houses, cars (with which you can come up with an interesting game), ending with complex models of radio-controlled or battery cars. By assembling models, children develop hand motor skills. This is especially useful for babies from 3 to 7 years old, so sometimes doctors even recommend modeling as a remedy.

In conclusion, we can say that modeling is incredibly exciting and even too much. Adherents of this hobby say that nothing can compare with the pleasure of creating a new world with your own hands. And participation in competitions, with independently made driving, floating and flying equipment, will delight any child. Modeling is a hobby that helps in understanding the world around us!

And in order to better understand whether this direction is interesting for you and your child, it is better to see with your own eyes the works of children who are fond of modeling. Come to the Siberian Gathering of Modelers, which will be held this weekend on September 22-23, on the second floor of the Pioneer shopping and entertainment center. Here you will find many impressive and memorable exhibits, as well as master classes where you can try yourself in this exciting process of creation.