11月7日下午,在信息化百人会2022年峰会主论坛上，信息化百人会学术委员会主席、工业和信息化部原副部长杨学山做了题为《制造企业数智化转型的基础模型及其分析》的主题演讲。杨学山指出，数字化转型尚处于发展早期，理论和实践尚未成熟。他通过分析数字化转型的相关模型，分享了他对制造企业数智化转型的相应看法。
他指出，在制造业数字化转型的过程中必须关注和克服“见数不见物、见物不见人”的问题，对制造业来说，不能忘记“物”与“数”之间的关系。任何数控装备、自动化生产线，包括物流的智能，都是“物”和“数”相连对制造业才有作用，否则没有用。同时，制造业的商务和决策阶段都离不开人的元素，我们不能只看到算法智能，忘记机器智能和人的智能。人的智能既是管理决策的基础，也是算法智能和机器智能的基础。
杨学山认为，制造企业的数字化转型是一个动态的过程，起点、项目、重点可以有所不同，但目的都是实现企业发展，方法就是快速迭代，在这两个问题上不存在差异，谁也不可能一步到位。
以下为杨学山发言内容整理：
我就制造企业的数智化转型模型问题谈一点看法，与大家分享。
先交待三件事情。一是为什么讲制造业数字化转型？大家知道我们在制定“中国制造2025”的时候，明确了制造业的发展方向是智能制造，或者说制造业朝向2025或2050，智能制造就是我们的方向。在这个过程中，我们一定要使用更多的数字和数字技术，所以说数字化转型和智能制造的方向合在一起就变成了数智化。 数字化转型网www.szhzxw.cn
二是数字化转型究竟处于什么状态？数字化转型，应该说还处于发展的早期，理论和实践都没有成熟，这是一个基本的判断。我们做任何事情一定要看看所处的方向和位置。为什么这样说？右边这个图，文字和这段话是来自德国工业4.0的那位著名专家的《工业人工智能》一书，我一字不动把它放过来，我们要想一想，GE做数字化转型是怎么推进的？它的力度有多少？最后是这样的评价，所以我们说数字化转型处于发展的早期是有依据的。
三是为什么要讲模型？因为模型可以分析一个企业和一个行业数字化转型的发展究竟是什么状态。同时，有了模型，也可以引导我们去做一个个具体的数字化转型实践。 数字化转型网www.szhzxw.cn
上面几个模型对于做数字化转型研究的机构或数字化转型比较先进的大企业来说都不陌生，这是该领域国际上几个知名机构和企业在总结全球数字化转型的企业做了什么，取得了什么成果，要朝着什么方向发展而总结出来的模型。 数字化转型网www.szhzxw.cn
对于这些模型，最近几年我不断地看、不断地想，我感觉它的归纳、趋势、一些方向很有道理。但是，我们很难按照它去进行评价和实践。为什么？原因很简单。回到刚才这页PPT，我们看GE，如果把GE做的事情和做的措施，按照这些评价模型来评价，它一定是很符合，评价很好，最顶尖的，但是结果不行，这就是很有意思的现实。这个现实说明，正如我刚才所说的数字化转型确实处于发展的早期，理论和实践还不够成熟，这是国外的几个例子。
我们再看国内的例子，这是数字化转型百问联合工作组在书里归纳的一个模型——工业互联网的参考体系结构模型。即使把“工业互联网”这个词换成“企业的数字化转型”或者“数字化转型”，我们发现居然什么都不用调整，模型也是OK的。这说明另外一个问题，我们今天很多新的概念，对于企业的生存和发展来说，对于企业的实践来说，其实是一回事。即使我们的学术界、研究机构需要去做多方向的研究和讨论，其实到了企业是一个事情。 数字化转型网www.szhzxw.cn
那这个模型相对于上面的模型有什么不同呢？一个重要的不同是它解剖了一个企业。无论是数字化转型、产业互联网、还是工业4.0，叫什么都没有关系。它给出了三个维度，其实主要是两个角度。一个角度是企业的发展过程分别从技术实现、功能、使用和业务4个视角来分析。同时，有一个生命周期，最右侧还有一个个的行业。相对来说，这个模型对于我们分析和部署全局有好处，我们可以一眼看到它的全局。
上面那本书是去年出版的，下面说的这本书这个月应该能出来，这是我今天要着重讲的一个模型。这个模型不是我的，这是赵敏他们三个人在《人本：从工业互联网走向数字文明》这本书里讲的工业互联网的生态模型。这个模型为什么要着重推荐和着重讲？我感觉这是迄今为止，从制造业的角度来看，不管是产业互联网、智能制造、数字化转型还是数智化转型，目前为止，见到的分析得最好的模型。
为什么是分析得最好的模型呢？原因是，这是真正懂制造业的人在沿着我们“中国制造2025”的方向往前走的时候，针对“需要分析什么？”“要做什么？”“这里面哪些因素是极其重要的？”这些问题，在这本书里做了一个比较清晰的交待。 数字化转型网www.szhzxw.cn
我们先看三维中底下的“资源维”，当然之后名字也许会变。我们先看“资源维”是什么意思，在座的都是制造业的企业，“资源维”讲的就是制造业核心的内容——产品由人、机、料、法、环、测来完成，这是制造的产品和过程，缺一不可。“实体维”十分有意思，是指在制造企业的制造过程中，从供应链到本企业到后面的销售售后服务，所有过程中相关的实体企业，也就是说决定制造的全过程，是由一个个企业决定的，是由这些实体决定的。往上这部分讲的是信息。这6层他们是这样划分出来的，我们可以精简也可以增加，但这样做对制造业来说确实有道理。
为什么首先有物理层？因为我们在制造过程中，在制造管理过程中用的数据从哪儿来？是从制造过程来的。我们看看人、机、料、法、环、测，“机”产生数据，“料”产生数据，“工艺”产生数据，生产的环境和检测产生数据，所以数据是从物理层上生出来的，从传感器变成比特层，但这个比特层不是控制，“控制”是到控制单元控制，到控制单元的时候比特数据变成可以用来控制的数据。我们从一个个具体生产过程的控制，往上管理的时候它认为OK就变成信息层，信息的汇总变成知识层，这样的汇总变成了一个企业在管理和商务活动中用来决策的数据和知识，所以这样的维度有道理。
也就是说，我们看数据不是看泛泛的数据，我们所有制造过程的数据是在这样的过程中产生的，这样的过程中产生的数据才在制造过程各个环节中发挥作用。
当然，“实体维”同样在业务流程中产生数据并进入到价值中，“实体”、“资源”、“数智”这样三维六层，加上一个价值流。数字化转型，产业互联网做所有的事情需要达到两个价值流。两个价值流是两个方向，一个方向是满足客户的价值，第二个是要实现制造的批量生产。所以这样才有价值，这些维度的活动才有价值和意义。这本书大概这个月会出版，最晚年底肯定会出版。 数字化转型网www.szhzxw.cn
所以，我建议真正做制造业数字化转型的或做产业互联网、工业大数据、工业4.0等等，这本书需要认真看看。为什么着重推荐这个模型？因为制造企业数字化转型必须以制造本身为基础，必须从参与制造的各个企业、各个过程来进行分析。
我们把这个模型往下打开看一下，一个制造企业的主要过程与资源。这主要包括三个过程，一个是制造本身，也就是这个产品从供应链开始一直到销售的售后，这样的全过程，一定要记住这是制造企业的根本、根基。这些活动，也就是供应链、销售、售后服务，必须有相应的商务活动来支撑。所有企业之外的事项都是以合同为依据，都是商务活动的结果。企业与企业之间不管是物的流动、人的流动、资金的流动还是数据的流动，都是企业与企业之间的商务，没有商务活动，这个流动是缺乏法律依据的。市场经济的核心依据是合同，是契约。当然，所有的事情都要经由管理，没有管理是不行的，所有的企业老总、中层管理人员都离不开管理环节。 数字化转型网www.szhzxw.cn
我们需要有资本、有数字、有技术、有人员。在这样的过程中，每一个过程所使用的资本、数字、技术、人员都在增长和提升。数字在增长，技术在提升，人的能力在提升，而且资本也在提升，这是理解数字化转型的关键所在。
它既是使用过程，又是提升的过程，而且每一个过程都是不可缺少的相互关联的。大体上，我们从制造企业的全貌来看就是这样的一个全貌。 数字化转型网www.szhzxw.cn
但是，数字化转型只是所有活动当中的一个部分，所以用黄线把它圈出来。对于制造企业来说，数字化转型的最核心问题是制造过程，也就是，无论是对企业发展的作用，还是我们在这里面的投入，我们在选择项目优先级的时候，一定是把制造放在首位。
为什么？因为一个制造企业之所以能够在市场经济中生存下来，是因为企业的产品，而且是因为企业的物质产品，企业才能在社会中有存在的价值。所以，刚才所讲的产品、人、机、料、法、环、测等若干要素，每个要素进行数字化的转型是关键的。对于企业，我们经常看到讲数字化转型，数字化转型的装备的更新却往往不算在里面。
大家想一想，我们讲人、机、料、法、环、测的时候，对于制造业企业，“机”和“法”就是工艺和装备，这是决定一个企业能够生产什么产品的关键。装备和工艺决定生产什么产品，产品生产出来是什么性能，甚至连成本都决定了，所以数字化转型怎么能不针对这样的关键环节？制造从供应链开始，为什么是“料”，“料”是原料部件，从别的地方过来。 数字化转型网www.szhzxw.cn
我们极其重视管理的过程，在数字化转型中管理过程如何提升？我们经常讲数据中台，因为制造过程是一片数据，管理过程是一组软件或者一套软件，而制造过程和管理过程中数据之间的Gap怎么解决？我们叫数据中台。在讲各种概念的时候，一定要在这样的全貌中找到究竟是做什么？为什么？我们千万不要被概念所迷惑。从制造过程到管理过程的这一层叫什么名字，没有关系。功能就是一个，是使得制造过程的数据能够有效地为管理过程所用，叫什么概念、名字，没有关系。
我们千万不要忘记商务过程，因为企业的供应链、销售、维护客户关系，是商务过程中实现的。所以，我们支持它。数字化转型不仅是支持制造和管理，还要支持商务。因此，我们不能忘记管理、商务和制造。
我们看到今天的数字化转型中，有几个必须要关注和克服的现象。大体归纳成这么几类，第一类是“见数不见物，见物不见人”。对制造业来说，不能忘记“物”与数据之间的关系。我们想一想，任何数控装备，任何自动化生产线，包括物流的智能，都是“物”和“数”相连对制造业才有作用，否则没有用。不能“见数不见物，见物不见人”，有了数据忘了物忘了人。对于企业来说，是人在决策，商务是由人来进行的。我们讲智能的时候，往往记住了人工智能的“智能”，忘记了机器智能和人的智能。装备、工艺是机器智能，加上人工智能还要加上人的“智能”，制造业才能发展。 数字化转型网www.szhzxw.cn
我刚才为什么推荐这本书，这本书重要的概念是我们不能只是看到了算法智能，忘记了机器智能和人的智能。人的智能既是管理决策的基础，也是算法智能和机器智能的基础，我们不能只见一种智能。我们不能只看见算法、算力平台，而忘了问题、价值、成本。算法、算力平台很重要，但它只是工具，工具永远为企业所用，为人所用，这是我们要看到的。
同时，我们还要说，制造企业的数字化转型，是一个动态的过程，起点可以不同，项目可以不同，重点可以不同，但是它的目的和快速迭代的方法是一样的。目的就是企业的发展，方法就是快速迭代，这是没有差异的，谁也不能达到一步到位。要想“打开天窗说亮话”，对于制造业来说，想一步到位的事情就是引进新的装备、新的生产线，别的事情不要想一步到位。 数字化转型网www.szhzxw.cn
翻译：
Yang Xueshan: The intelligent transformation of manufacturing enterprises, this model must pay attention to!
On the afternoon of November 7, at the main forum of the 2022 Summit of the Information Committee of 100, Yang Xueshan, chairman of the Academic Committee of the Information Committee of 100 and former Vice Minister of the Ministry of Industry and Information Technology, made a keynote speech entitled “The Basic Model and Analysis of the digital intelligent transformation of Manufacturing Enterprises”. Yang Xueshan pointed out that digital transformation is still in the early stage of development, and the theory and practice are not yet mature. By analyzing the relevant models of digital transformation, he shared his corresponding views on the digital intelligent transformation of manufacturing enterprises. 数字化转型网www.szhzxw.cn
He pointed out that in the process of digital transformation of the manufacturing industry, we must pay attention to and overcome the problem of “seeing things, seeing things and not seeing people”, and for the manufacturing industry, we can not forget the relationship between “things” and “numbers”. Any CNC equipment, automated production lines, including the intelligence of logistics, are “things” and “numbers” connected to the manufacturing industry, otherwise it is useless. At the same time, the business and decision-making stages of the manufacturing industry are inseparable from the human element, we can not only see the algorithmic intelligence, forget the machine intelligence and human intelligence. Human intelligence is the basis of management decisions, as well as algorithmic intelligence and machine intelligence.
Yang Xueshan believes that the digital transformation of manufacturing enterprises is a dynamic process, the starting point, the project, the focus can be different, but the purpose is to achieve enterprise development, the method is rapid iteration, there is no difference in these two issues, no one can be in place in one step.
The following is the content of Yang Xueshan’s speech.
I would like to share with you my views on the digital intelligent transformation model of manufacturing enterprises. 数字化转型网www.szhzxw.cn
Three things first.
First, why the digital transformation of manufacturing? As you know, when we formulated “Made in China 2025”, we made it clear that the development direction of the manufacturing industry is intelligent manufacturing, or manufacturing towards 2025 or 2050, intelligent manufacturing is our direction. In this process, we must use more digital and digital technology, so the direction of digital transformation and intelligent manufacturing together becomes digital intelligence.
Second, what is the state of digital transformation? Digital transformation, it should be said, is still in the early stage of development, and both theory and practice are not mature, which is a basic judgment. We must look at the direction and position of everything we do. Why do you say that? The figure on the right, the text and this paragraph are from the famous expert of German Industry 4.0 “Industrial Artificial Intelligence” book, I put it over without moving a word, we have to think about, GE to do digital transformation is how to promote? How strong is it? Finally, this is the assessment, so we say that digital transformation is in the early stages of development is justified.
Third, why do we talk about models? Because the model can analyze the state of the development of digital transformation of an enterprise and an industry. At the same time, with the model, it can also guide us to do a specific digital transformation practice.
The above models are not unfamiliar to institutions that do digital transformation research or large enterprises that are relatively advanced in digital transformation.
This is a model summarized by several well-known international institutions and enterprises in the field to sum up what global digital transformation enterprises have done, what results have been achieved, and what direction to develop. 数字化转型网www.szhzxw.cn
For these models, in recent years, I have been constantly looking at and thinking about them, and I feel that its induction, trends, and some directions are very reasonable. However, it is difficult for us to evaluate and practice according to it. Why? The reason is simple. Going back to this PPT, let’s look at GE, if we evaluate what GE does and what it does according to these evaluation models, it must be very consistent, the evaluation is very good, the top, but the result is not good, this is a very interesting reality. This reality shows that, as I just said, digital transformation is indeed in the early stage of development, and the theory and practice are not mature enough. Here are a few examples from abroad.
Let’s look at the domestic example, this is a model summed up in the book of the digital transformation Joint Working group – the reference architecture model of the industrial Internet. Even if we replace the term “industrial Internet” with “digital transformation of enterprises” or “digital transformation”, we find that nothing needs to be adjusted, and the model is OK. This shows another problem, many of our new concepts today, for the survival and development of enterprises, for the practice of enterprises, in fact, the same thing. Even if our academic circles and research institutions need to do research and discussion in multiple directions, in fact, it is one thing to the enterprise.
So how does this model differ from the model above?
One important difference is that it dissects a business. Whether it’s digital transformation, industrial Internet, or Industry 4.0, it doesn’t matter what it’s called. It gives three dimensions, actually two main angles. One perspective is to analyze the development process of enterprises from four perspectives: technology realization, function, use and business. At the same time, there is a life cycle, and there are industries on the far right. Relatively speaking, this model is good for us to analyze and deploy the big picture, which we can see at a glance. 数字化转型网www.szhzxw.cn
The above book was published last year, the following book should be out this month, and this is a model that I will focus on today. This model is not mine, this is the ecological model of the industrial Internet described by Zhao Min and their three people in the book “People: From the Industrial Internet to Digital Civilization”. Why should this model be recommended and emphasized? I feel that this is by far the best analyzed model we have seen from a manufacturing perspective, whether it is industrial Internet, intelligent manufacturing, digital transformation or digital intelligent transformation.
Why is it the best analyzed model?
The reason is that this is the time for people who really understand manufacturing to move forward along the direction of our “Made in China 2025”, to “what needs to be analyzed?” “To do what?” “What factors are extremely important?” These questions, in this book made a relatively clear explanation.
Let’s first look at the “resource dimension” at the bottom of the 3D, of course, the name may change later. We first look at the meaning of “resource dimension”, all of you are manufacturing enterprises, “resource dimension” is the core content of manufacturing – products by people, machines, materials, law, ring, measurement to complete, this is the manufacture of products and processes, indispensable. “Entity dimension” is very interesting, refers to the manufacturing process of the manufacturing enterprise, from the supply chain to the enterprise to the subsequent sales and after-sales service, all the relevant entities in the process, that is to say, the whole process of determining the manufacturing is decided by one enterprise, is decided by these entities. The upper part is about information. These six layers are divided this way, we can streamline or increase, but it does make sense for the manufacturing industry.
Why is there a physical layer in the first place?
Because where does the data that we use in the manufacturing process, in the manufacturing management process come from? It’s from the manufacturing process. We look at human, machine, material, method, ring, measurement, “machine” to produce data, “material” to produce data, “process” to produce data, production environment and detection to produce data, so the data is born from the physical layer, from the sensor to the bit layer, but this bit layer is not control, “control” is to control the control unit.
By the time it reaches the control unit, the bit data becomes data that can be used for control. When we go from the control of a specific production process to the upward management, it thinks that OK becomes the information layer, the summary of information becomes the knowledge layer, and such summary becomes the data and knowledge used by an enterprise to make decisions in management and business activities, so this dimension makes sense. 数字化转型网www.szhzxw.cn
In other words, we do not look at the data to see the general data, the data of all our manufacturing processes is generated in such a process, and the data generated in such a process plays a role in all aspects of the manufacturing process.
Of course, the “entity dimension” also generates data in the business process and enters the value
Of course, the “entity dimension” also generates data in the business process and enters the value, such as the three-dimensional six layers of “entity”, “resource” and “digital intelligence”, plus a value stream. Digital transformation, the industrial Internet to do everything needs to reach two value streams. The two value streams are two directions, one direction is to meet the value of the customer, the second is to achieve the mass production of manufacturing. So this is valuable, these dimensions of activity have value and significance. The book will be published probably this month, certainly by the end of the year at the latest.
Therefore, I suggest that we really do the digital transformation of the manufacturing industry or do the industrial Internet, industrial big data, industry 4.0, etc., this book needs to be carefully looked at. Why is this model highly recommended? Because the digital transformation of manufacturing enterprises must be based on the manufacturing itself. It must be analyzed from the various enterprises and processes involved in manufacturing. 数字化转型网www.szhzxw.cn
Let’s take this model down and look at the main processes and resources of a manufacturing company.
This mainly includes three processes, one is the manufacturing itself. That is, the product from the supply chain to the after-sales sales, such a whole process. We must remember that this is the foundation of manufacturing enterprises. These activities, namely supply chain, sales, after-sales service, must be supported by corresponding business activities. All matters outside the enterprise are based on contracts and are the result of business activities. Whether it is the flow of things, people, funds or data between enterprises. It is the business between enterprises, without business activities, this flow is lack of legal basis. The core basis of market economy is contract, contract. Of course, all things have to go through management, no management is not possible, all corporate managers, middle managers are inseparable from the management link.
We need capital, we need numbers, we need technology, we need people. In such a process, the capital, numbers, technology, and people used in each process are growing and improving. Digital is growing, technology is improving, human capabilities are improving, and capital is also improving, which is the key to understanding digital transformation.
It is both a process of use and a process of promotion, and each process is indispensable and interrelated. In general, we from the overall picture of manufacturing enterprises is such a picture.
However, digital transformation is only one part of all activities, so circle it with the yellow line. For manufacturing companies, the most central issue of digital transformation is the manufacturing process. That is, whether it is the role of enterprise development, or our investment in this. We must put manufacturing first when choosing project priorities.
Why? Because the reason why a manufacturing enterprise can survive in the market economy is because of its products. And because of the material products of the enterprise, the enterprise can have the value of existence in the society. Therefore, the product, human, machine, material, method, environment, measurement and other elements just mentioned, each element of the digital transformation is key. For enterprises, we often see digital transformation, but the update of digital transformation equipment is often not included. 数字化转型网www.szhzxw.cn
Think about it, when we talk about people, machinery, materials, law, environment, and measurement. For manufacturing enterprises, “machine” and “law” are processes and equipment, which is the key to determining what products an enterprise can produce. Equipment and processes determine what products are produced, what performance the products are produced. And even the cost is determined, so how can digital transformation not target such key links? Manufacturing starts from the supply chain, why is “material”, “material” is the raw material parts, come from somewhere else.
We attach great importance to the management process.
How can the management process be improved in digital transformation? We often talk about the data center. Because the manufacturing process is a piece of data, the management process is a group of software or a set of software. And how to solve the Gap between the manufacturing process and the management process data? We’re called data center. When talking about various concepts, we must find out what we are doing in this overall picture. Why? We must not be confused by concepts. It does not matter what the name of this layer from the manufacturing process to the management process is. Function is one, is to make the data of the manufacturing process can be effectively used for the management process, what is the concept, the name, it does not matter.
We must not forget the business process. Because the supply chain, sales, maintenance of customer relations, is achieved in the business process. So, we support it. Digital transformation supports not only manufacturing and management, but also commerce. So we can’t forget management, business and manufacturing. 数字化转型网www.szhzxw.cn
We see several phenomena in today’s digital transformation that must be addressed and overcome.
Generally summed up into such a few categories. The first category is “see the number of things, see things do not see people.” For manufacturing, the relationship between “things” and data cannot be forgotten. We think about it, any CNC equipment, any automated production line. Including the intelligence of logistics, is “thing” and “number” connected to the manufacturing industry, otherwise it is useless. Can not “see the number of things, see things do not see people”, with data forget things forget people. For enterprises, it is people who make decisions and business is carried out by people. When we talk about intelligence, we often remember the “intelligence” of artificial intelligence, forgetting machine intelligence and human intelligence. Equipment and technology are machine intelligence, coupled with artificial intelligence and human “intelligence”, the manufacturing industry can develop.
The important concept of this book is that we can’t just look at algorithmic intelligence and forget about machine intelligence and human intelligence.
Human intelligence is the basis of management decisions, algorithmic intelligence and machine intelligence, and we cannot see only one kind of intelligence. We can’t just look at the algorithm, the computing power platform, and forget the problem, the value, the cost. Algorithm, computing power platform is very important. But it is just a tool, the tool is always used by the enterprise, used by people, this is what we want to see. 数字化转型网www.szhzxw.cn
At the same time, we also want to say that the digital transformation of manufacturing enterprises is a dynamic process. The starting point can be different, the project can be different. The focus can be different, but its purpose and rapid iteration method is the same. The purpose is the development of the enterprise, the method is rapid iteration, which is no difference. No one can achieve one step in place. In order to “open the skylight to make it clear”, for the manufacturing industry. The thing that wants to be in place in one step is to introduce new equipment and new production lines. And other things do not want to be in place in one step. 数字化转型网www.szhzxw.cn
本文由数字化转型网（www.szhzxw.cn）转载而成，来源于信息化百人会 ；编辑/翻译：数字化转型网宁檬树。
免责声明: 本网站(https://www.szhzxw.cn/)内容主要来自原创、合作媒体供稿和第三方投稿，凡在本网站出现的信息，均仅供参考。本网站将尽力确保所提供信息的准确性及可靠性，但不保证有关资料的准确性及可靠性，读者在使用前请进一步核实，并对任何自主决定的行为负责。本网站对有关资料所引致的错误、不确或遗漏，概不负任何法律责任。
本网站刊载的所有内容(包括但不仅限文字、图片、LOGO、音频、视频、软件、程序等) 版权归原作者所有。任何单位或个人认为本网站中的内容可能涉嫌侵犯其知识产权或存在不实内容时，请及时通知本站，予以删除。
