In the tide of intelligent manufacturing, China actively promotes the construction of industrial Internet, and manufacturing industry is also actively transforming. The level of automation, digitization and intelligence has been greatly improved. Leading enterprises in various industries pay more and more attention to the data collection and workshop networking of production equipment, and the visibility of equipment data has been significantly improved. However, although most manufacturing enterprises spend a lot of money on very advanced equipment, equipment management and maintenance, personnel knowledge structure still stay at a lower level: data acquisition is basically pen based, processing is basically human, analysis is based on guessing (experience). It can be said that the equipment of industry 4.0 was bought, but it continued the management of industry 2.0.
From the perspective of the benefit output of equipment assets, there is a huge space for the improvement of the comprehensive efficiency of equipment in China's manufacturing industry. According to statistics, OEE of most discrete manufacturing industry in China is about 40%, and there is still room for improvement of at least 30% - 40% from developed countries. At the same time, many enterprises pay little attention to the fine management of equipment maintenance and maintenance, which leads to hidden losses such as abnormal shutdown and spare parts waste. Under the situation of the current epidemic impact and the uncertainty of globalization situation, improving the level of equipment management can bring valuable opportunities for the survival and development of enterprises and the promotion of competitiveness.
How to build the management ability of new equipment is a problem and challenge facing the manufacturing industry in China. Therefore, we summarize the five mistakes in manufacturing enterprises, and give corresponding strategies and suggestions, hoping to help enterprises to take some detours.
01
Emphasis on hard but not soft
Most enterprises only pay attention to the acceptance and handover of hardware, ignore the operation, maintenance and service standards of software system, and do not explicitly require equipment manufacturers to provide data acquisition interface and define equipment data ownership.
According to relevant data statistics, at present, the digitization rate of production equipment in China's enterprises is 47%, the numerical control rate of key processes is 51%, and the networking rate of key equipment is 41%. Embedded software, human-computer interface, data monitoring model and management platform are important components of intelligent devices, and they should also be the category of device management. Combined with the author's experience in industrial Internet related projects, equipment data acquisition is still one of the biggest pain points in the promotion of digitalization of production site due to many industrial site protocols, the original factory is not open and does not support, and equipment data is not confirmed.
For example, in many SMT production lines in China, the mounter itself has high precision, fast beat, and the yield rate is more than 99%. It is difficult to improve simply by manual debugging. Many factories expect to be able to collect and analyze the data of the placement machine in real time, solve the problem of automatic material calling and receiving, and improve the problem of material throwing. But at present, tens of thousands of data acquisition license fees make many factories flinch.
Therefore, the factory will consider in advance in the equipment procurement link, and add the relevant requirements into the commercial terms, which can prepare for the future equipment manufacturing process detailed data collection and process, quality analysis and optimization.
02
Production is the most important, no matter whether it is broken or repaired or not
In most factories, especially in the discrete manufacturing industry, production is the leader, and the equipment is only the guarantee department. As long as the equipment can operate, it will not stop production. This leads to a serious shortage of time and capital investment for equipment maintenance and improvement, and the equipment department also falls into a passive vicious circle of fighting fires everywhere and struggling to cope. The reason is that the enterprise does not look at the loss of equipment downtime from the end-to-end perspective of the factory. When the initial symptoms of equipment failure appear, maintenance is far less than the loss and input cost caused by maintenance after shutdown.
As shown in the figure below, through vibration analysis of a machine tool, it can be seen that the peak vibration acceleration on October 18 triggered the early warning threshold, but due to production planning problems, there was no downtime for maintenance; On October 22, the main control system of the machine tool broke down and had to stop for 10 hours to repair and replace the bearing. After the repair, the vibration returned to normal level, but a large loss had been caused.
Figure 1: most mechanical faults can be monitored by vibration analysis
Equipment management has experienced four development processes: from corrective maintenance (CM) of 1.0 to preventive maintenance (PM) of 2.0, reliability maintenance (RCM) of 3.0 and predictive maintenance (PHM) of 4.0. Essentially, equipment health management (EHM) is an evolutionary process from 'treating the disease' to 'preventing the disease'. Through EHM, the equipment health status is no longer simply divided into normal and abnormal.
We can evaluate the sub-health status of the equipment through new technology, new tools and analysis of the accumulated basic data, so as to maintain the equipment in advance and greatly reduce the cost of equipment maintenance. For example, gechuangdongzhi ehm equips equipment maintenance engineers with intelligent spot detector with vibration sensor, just like equipping doctors with 'intelligent Stethoscope'. By monitoring vibration for a few seconds, combined with the built-in spectrum analysis model, it can accurately and quickly judge the health status of equipment and the cause of fault symptoms, which plays an important auxiliary role in fault diagnosis of equipment engineers. In this way, the responsibility of the equipment management personnel has changed from the original repair work to the professional maintenance work of how to ensure the healthy operation of the equipment, and entered a virtuous circle.
03
The equipment problem is the equipment department
Equipment management has experienced four development processes: from corrective maintenance (CM) of 1.0 to preventive maintenance (PM) of 2.0, reliability maintenance (RCM) of 3.0 and predictive maintenance (PHM) of 4.0. Essentially, equipment health management (EHM) is an evolutionary process from 'treating the disease' to 'preventing the disease'. Through EHM, the equipment health status is no longer simply divided into normal and abnormal.
We can evaluate the sub-health status of the equipment through new technology, new tools and analysis of the accumulated basic data, so as to maintain the equipment in advance and greatly reduce the cost of equipment maintenance. For example, gechuangdongzhi ehm equips equipment maintenance engineers with intelligent spot detector with vibration sensor, just like equipping doctors with 'intelligent Stethoscope'. By monitoring vibration for a few seconds, combined with the built-in spectrum analysis model, it can accurately and quickly judge the health status of equipment and the cause of fault symptoms, which plays an important auxiliary role in fault diagnosis of equipment engineers. In this way, the responsibility of the equipment management personnel has changed from the original repair work to the professional maintenance work of how to ensure the healthy operation of the equipment, and entered a virtuous circle.
04
Regard equipment maintenance as cost center
Ignoring the loss under the iceberg
Although TPM has been implemented for many years, many managers still think that there is a problem with the equipment department, which leads to the production department not caring about the equipment failure and neglecting the maintenance of the equipment which affects the production and quality. Equipment maintenance engineers often laugh at the watchdog and scapegoats because of their low status and low salary: on holidays, others can rest, but they can not leave, because this is a good time to repair the equipment; Any problem, whether it is equipment shutdown, production shutdown, or quality accidents, will be related to the equipment, equipment personnel are almost always the back of the pot man. The equipment department has become the place where excellent talents are the least willing to go. This vicious circle phenomenon requires the production managers to work hard, establish the correct equipment management concept and build the production as the main body of the independent maintenance system. Only when the managers of production department attach importance to it, the operators of production equipment will change their indifferent attitude to the equipment, and can the equipment maintenance be carried out effectively. This equipment department alone is not going to play.
The factory can build the system and operation mechanism of the whole staff independent maintenance by introducing the digital ehm solution of equipment health management. For example, an electronic factory, through the introduction of EHM, realized the secondary standard maintenance system of equipment: Daily independent spot inspection and maintenance of station operators, professional spot inspection and maintenance of equipment engineers, and the elimination of the original problem of false spot inspection and false inspection through NFC, mobile app, photo watermark and image comparison and peer-to-peer technology. At the same time, the functions and mechanisms of scanning code repair, automatic repair of data rules, self robbing of maintenance work order, user evaluation of maintenance effect, and competition ratio of performance score are used to help the factory realize the self operation of TPM. Half a year or so, it helps the factory reduce the abnormal downtime by about 20%.
Figure 2: the essence of equipment management is to find the best maintenance balance
05
Want to rely on predictive maintenance to solve problems
Ignoring the basic digital construction and data accumulation
'Predictive maintenance' has always been a hot topic in the industrial Internet. Many companies claim to achieve predictive maintenance through IOT and AI. Many factories also expect to hand over their uncertainty about equipment failure to 'predictive maintenance'. However, according to the observation, the accuracy of most of these projects is very low, and most of them are conceptual and experimental. There are still problems in the interpretability, verifiability and replicability.
The implementation of predictive maintenance is more difficult than expected because it is far more difficult to extract the interpretable logic of industrial mechanism by relying solely on data.
There are two main reasons: first, the basic data of many enterprises are still lack of accumulation. For example, the basic patrol inspection, maintenance and fault analysis records of equipment are still scattered in all kinds of paper and excel. The equipment lacks digital archives, basic maintenance data, spare parts replacement records, fault and repair data, It is impossible to train and verify the model without structured accumulation of fault feature data; Second, many manufacturers try to rely solely on data analysis path and ignore the integration of existing professional knowledge and experience of equipment engineers. Relying on Mathematics and AI algorithm alone, they are easy to fall into the statistical trap, only get the correlation, and it is not easy to get an interpretable and predictable causal model.
Therefore, we suggest that factories: first, they should pay attention to the construction of basic digital capabilities such as equipment digital archives, basic maintenance, repair work orders, fault trees, etc. The second is to combine the empirical model with the data model for the key high-value and high loss equipment, and the output of the model is to assist the maintenance and repair of personnel. Finally, it needs to be handed over to people for comprehensive judgment.
In general, equipment is to a factory what a gun is to a soldier. Many equipment maintenance and repair technology systems are indeed developed from the military weapon maintenance system. To build a new type of equipment management capability, factory managers need to recognize that equipment is the basis of building the core competitiveness of the factory, actively change the equipment management and operation mode, and develop to digital and intelligent. According to Gartner's prediction, by 2022, more than 60% of the equipment will realize the data-based intelligent operation and maintenance mode, and the intelligent management and operation and maintenance ability of the equipment will be an important symbol to measure the core competitiveness of a factory.
In order to promote the construction of new capacity of plant equipment management, the relevant national departments are also drafting and formulating the national standards of equipment management and the framework of capability maturity assessment, which will play an important role in guiding and promoting enterprises to strengthen the capability of equipment management.
The article is from the official account of the national science and Technology Corporation.