Production line layout is actually code jumping

The placement of machines in a factory is a university. It needs to think about business logic, logistics and production process from the overall situation of the factory level. With the help of plant layout or production planning software, users can model the material flow of processes at all levels, so as to realize plant level optimization. At this moment, we will also look for the best production and manufacturing strategy for the global networked production facilities.

Here, it is necessary to design the large process of the factory, including the layout of machine production line, the setting of gate, the access channel of logistics, the layout of line side warehouse, etc. Of course, it also includes the design of people's access. The layout of machines and shuttle feeding transportation equipment is closely related to the process route of the whole production line.

If the same injection molding machine is placed in different workshops or even entered into different workshops, its layout will change completely. More importantly, planners need to simulate and verify various processes and assembly planning, simulate the whole production and manufacturing process, get the verification results before putting into production, and generate lists, reports and instructions.

The process simulation of this production line obviously needs to have a better understanding and data source of the machine itself. At the end of last century, many professional simulation companies such as technomatix in Israel and Deneb robotics in the United States entered the field of vision. The automotive industry has loved this and adopted it one after another, which has greatly promoted the development of this kind of digital production line process simulation software. Because the machine data needs to come from the previous machine design data, the CAD manufacturers of designing machinery quickly realized that this is an indispensable position, so the traditional mechanical design CAD manufacturers entered one after another. The two manufacturers mentioned above were also acquired respectively. CAD manufacturers also left the boundary of the original design department and began to wander between the production workshop and the plant.

Very interesting thing is the simulation of personnel flow, such as anylogic simulation software. This company for pedestrian simulation is worth preserving, because finally we can see a software from Russia in the industrial field. It can realize the modeling and Simulation of system dynamics, discrete events and agents. It is widely used in airports, museums, subways and other crowded places, which require a deep understanding of human behavior models. People's walking will be affected by self motivation mechanism, such as always trying to surpass the people in front of them. This sounds simple, but things on the factory site are much more complicated.

The sudden increase in orders for complex products is often the honey of distressed people. For example, the production increase of aircraft is a complex process, because it needs to reassess the impact of the production line. Due to the low output, many work units are generally customized, and the parts are produced in different regions, so the anti-interference ability is actually very poor. These factors will lead to the increase of aircraft production cycle when expanding production.

In order to cope with this change, the EU launched the "adaptive production management project", focusing on increasing the continuous growth capacity of new products in the aircraft and shipbuilding industry. Airbus also joined. The assembly line of A350 in Hamburg (commonly known as pulse production line in the industry) will complete the assembly of two different parts of the fuselage. It contains 6 assembly stations, each station includes 30-35 employees and about 300 work orders. The productivity of this production line will increase over time, and the whole production cycle will last for two years. Here, each workstation is regarded as an agent. Then, how to simulate this process, how to modify the operation of personnel and the issuance of event instructions, so as to adapt to the changes brought by increasing production.

     
 

Figure 1 adaptive management of Airbus

As a part of adaptive production management, there must be simulation software to enable participants to reproduce the real production environment. Anylogic was successfully selected for its hybrid modeling advantages based on agents and discrete events. In fact, it also has many production scheduling, WIP strategy management and other software. There will be many strange scenes and process requirements, so there will be unexpected software to meet the needs.

In addition to anylogic, other personnel simulation software includes step and massmotion from the UK. It is worth mentioning that the latter participated in the passenger flow simulation of Beijing Daxing airport, which provides services for 45 million passengers every year. Personnel flow and emergency evacuation must be considered.

At the same time, material flow also needs simulation. In the lean system, material is the most important link. There is a profound efficiency password behind the location and stock of materials. It can be said that materials are the most conspicuous scale of waste. The logistics in a factory often carries out interactive throughput with materials and equipment such as three-dimensional warehouse, freight car, forklift, pallet and so on through the warehousing and logistics software WMS. When the space of raw materials, semi-finished products, parts and work in progress is greatly moved, it is dominated by a set of strict production plan information flow.

So, how did this daily torrent of goods change from imagination to reality?

The logistics design of an enterprise can often be invited by the owner to a professional logistics consulting company or by the design institute. The design core of various materials is actually the simulation of discrete events. The most commonly used AutoMod software also includes American Flexsim software and ProModel. Flexsim is widely used in large-scale warehousing and logistics (such as JD's three-dimensional warehouse). In some relatively simple factories, it has many applications and is easy to use. The large equipment manufacturing industry also shows a preference for software. It is hard to imagine that AutoMod, a logistics simulation software, is actually the software of Yingcai, the largest semiconductor equipment manufacturer in the United States. As early as 2006, Yingcai purchased this logistics simulation company from an automation company at a cost of US $125 million, so as to accelerate the support of flexible productivity in the semiconductor industry and avoid expensive and highly customized production line solutions. It is not surprising that equipment manufacturers have commercial software, and Mitsui shipyard has developed a similar logistics simulation software. This is also a feature of Japanese software development. In the vertical field, there will always be unexpected professional software.

Figure 2. Some logistics simulation software

Of course, large professional industrial software companies will also provide logistics system simulation and analysis tools, such as Dassault DELMIA quest, Siemens plant simulation and KUKA visual components. As a logistics simulation software, quest was one of the dowries obtained by Dassault system when it acquired Deneb company in 2000. Siemens plant simulation is the same story. When its UG acquired technomatix, EM plant was evolved.

For automation suppliers, automation and machines need to be more closely combined, so they will have greater enthusiasm to promote the integration of this software and automation system. This is why as early as 2000, Rockwell acquired Arena software with modeling and simulation as its specialty, and 55 employees were incorporated into the software department of Rockwell Automation Company, which has been retained until now. Now, in the first year of the 21st century, there were three less significant M & A events (there should be more), all related to factory production line simulation. This shows that the logistics simulation system has firmly become a part of the digital production line at the beginning of the century. Since then, the relationship between automation and software began to become closer. The faint sound of this budding Festival finally banged with the help of the globalization of German industry 4.0 15 years later, and every factory can hear it.

If some planners are enthusiasts of human factors engineering, they need to design the station more complex with human nature and reduce the sense of fatigue. You can also use ergonomic software, a small software of a German company, for station evaluation.

From the perspective of implementation, the layout of machine production line needs to be combined with another broad and profound in plant management: Industrial Engineering. Over the years, as a fruit of the combination of automation and industrial engineering, the optimized layout of production line has been mature and has basically become the standard configuration. However, with the development of digital technology including the Internet of things, there are more powerful tools for the layout and Simulation of machines, stations, materials and personnel. The pioneer of digital factory wants to do too much.

Industrial software is like a leisurely light. It climbs down from the ceiling of the factory and flows through production lines. With the flow of materials in the three-dimensional warehouse, it permeates the whole factory. After the modeling of the plant and production line is completed, you can enter the link of machine simulation and debugging.

Machine debugging, can you be faster

When the machine is transported into the plant, it is like a melody already arranged in the digital space. Each machine is like a note, which only needs to be placed at the corresponding position of the staff. Before each small paragraph, whether it is a weak P or a strong F has been carefully polished by the composer. After taking root, performers need to start trying to take out their own instruments for ensemble. This is a time for on-site machine commissioning, including the running in of automation, electrical and mechanical devices. Various configurations and machine joint debugging until the machines on the whole production line roar. However, the machine debugging in the factory is far from as pleasant as the performance of the movement. On the contrary, it is a painstaking task. How many excellent engineers left tons of sweat here. Many seemingly unbreakable knots have become one of the nightmares of field engineers.

How to make the on-site debugging shorter? The best way is that a production line and a machine are like a USB plug-in disk, which can be downloaded and used. This ideal plug-in factory has become the highest level of production line simulation. To achieve this, it is necessary to add specific parameters of various equipment to simulate the real-time scene during the previous production line layout. It makes the control system PLC or the nameplate parameters of a cable into boundary conditions and inputs them into the simulation system. In a word, production lines, like products, are designed precisely. This is a strange environment for traditional automation manufacturers. As the control system and connector of the machine, and the driver of materials, automation is expected to take a step forward. The digital age has exacerbated this trend. In the digital age, if there is no such production line simulation technology, it will be lame.

After Siemens acquired PLM manufacturers such as UG in 2006, it had the inherent advantage of simulation of production line layout. However, it further extended to machine virtual mode. This puts great pressure on other automation manufacturers, especially automation companies with relatively small volume such as Rockwell. In the last three years, Rockwell has a two-step layout worthy of attention. The first step is to invest US $1 billion in PTC, which is almost equivalent to the annual output value of PTC in exchange for a strategic binding. Strengthen the Internet of things connectivity of machines and equipment, and also strengthen the connection with other machine data; The second step is to acquire emulate3d, an engineering software developer engaged in digital simulation, in early 2019. The investment in this kind of M & A is not large, but it has profound significance for machines to enter the digital space. The software can establish a simple factory model, but its focus is to connect with the control system to realize virtual control. In other words, although it can not do process simulation, it can do control based simulation. As a PLC simulation program, emulate3d is the short board that rockwell automation needs to supplement. It can establish a simulation space for virtual debugging for PLC and other control systems and view the program logic and results. Interestingly, this company was originally a partner of Rockwell. It's too common to eat small partners in the field of software ecology. This kind of close relative marriage is also a natural thing.

There are also some independent manufacturers, such as industrial physics. On the one hand, it can directly import CAD model, and at the same time, it can also virtual test PLC to form the digital twin of PLC. How long such a company can be independent is also a question mark. It may soon be eaten by automation companies.

Finally, the results of virtual debugging will be directly transferred to the controller as real code. The commissioning time on site will be greatly shortened. This means that automation manufacturers have entered the digital production line by expanding the controller simulation software. This is also an important basis for the development of digital twins of each machine. Without effective control means, the digital twin is a puppet without a hanging line.

A welding equipment engineering company in Sichuan provides automatic welding lines for automobile bodies. A production line project in a European automobile factory has realized the virtual debugging of a comprehensive digital production line. During commissioning and remote installation guidance, the equipment is directly transported to the plant area and a successful commissioning is achieved, which greatly shortens the commissioning time. The digital twin of production line has a more and more solid foundation.

In the process industry, this kind of virtual debugging is very common. For example, Schneider Electric aveva virtual debugging and operator training system (OTS) based on dynsim is to adjust the parameters of distributed control system DCS in the design stage. The adjusted parameters can be directly distributed to DCS system. In fact, such a software plays a variety of roles. Its most important work is dynamic process simulation, which simulates dynamic processes such as temperature and flow changes in advance. This technology is used as the computing kernel engine of simulation training, and the developed operator training system OTS is also widely popular. After the emergence of DCS in that year, in order to adapt to this control system, personnel need to be trained in advance. Therefore, OTS is also popular. The emergence of OTS also promotes the development of chemical process dynamic simulation software.

China has taken many detours in the control and optimization of chemical industry. According to Mr. Xu of China Chemical Information Center, in the late 1980s, the Ministry of chemical industry at that time had begun to carry out the research on plant control operation optimization. Nanjing Chemical Research Institute and the computing center of the Ministry of chemical industry have participated in the development of this software. However, due to the constraints of computer hardware, the data are analyzed offline. Finally, it is found that these software and algorithms have no practical value. Even if the optimized process operating conditions are calculated from this data, the operating conditions of the actual plant have already changed. It is impossible to infer the actual operation simply by mathematical model.

This lesson is still of great significance today. It is only limited to the analysis of big data and the regression calculation of mathematical model, and the extension error will be great due to the lack of research on mechanism model. The focus of tackling difficulties lies not in the mathematical model of big data, but in the industrial mechanism itself.

Notes: penetration between miracles and reality

From factory construction, to production line layout, to machine debugging, all kinds of industrial software are like Wu Yong, a smart star in our mind, who can properly arrange virtual people, machines and materials in digital space. When the virtual verification is completed, the production activities of an enterprise officially begin, and the previous deduction is not known. Like the film "wonderful night at the museum", when the sun rises and people enter the scene, they don't know that there was a thrilling confrontation and compromise here the night before. If you don't understand the darkness of night during the day, it's difficult to understand the digital heart in physics. Industrial software that can't appear is a cover between miracles and reality. If you don't understand it, you can't understand industrial intelligence.

Disclaimer: This article comes from the InternetReprinted from“Knowledge automation”, support the protection of intellectual property rights. Please indicate the original source and author for reprint. If there is infringement, please contact us for deletion.