Recently, the Ministry of Industry and Information Technology “Industrial Internet Innovation and Development Action Plan (2021-2023)” proposed that by 2023, the construction of new industrial Internet infrastructure will advance in quantity and quality, new models and new formats will be widely promoted, and the comprehensive strength of the industry will be significantly improved.

BL0170-PC0010-P At the policy level, industrial Internet policies continue to land, in fact, to the enterprise level, industrial Internet application is a different picture. Let’s start with two data, Deloitte’s “Artificial Intelligence Manufacturing Application survey” data shows that although manufacturing enterprises implement artificial intelligence projects, only 9% of implementation enterprises think that the project reaches their expectations of 80-100%; During the two sessions, Xu Xiaolan combined with previous research data said that China has not yet completed the basic equipment digital transformation of more than 55% of the enterprises.

Especially for small and medium-sized enterprises, BL0170-PC0010-P data has become the first step in their transformation and upgrading. Looking at all kinds of industrial software, such as the MES system used in manufacturing, the PLM system used in managing the whole life cycle of products, and the CRM system used in sales and service, their basic steps all start from the data. No matter what kind of digital system, data is the foundation, the formation of data closed-loop flow is the key. This is based on data that is efficiently collected, transmitted, stored, and analyzed.

BL0170-PC0010-P

BL0170-PC0010-P In fact, enterprise digitalization pays more attention to the creation of software systems, often ignoring hardware facilities, which is precisely an important step related to the source of data. Especially in manufacturing enterprises, a large amount of data comes from the equipment layer, and data acquisition is the key. This data acquisition process includes devices, sensing devices, transmission networks, artificial intelligence, industrial big data, and cloud computing.

At the device level, we first consider what data needs to be collected from which devices. This part involves the overall planning and system construction of enterprise digitalization. Adjust according to the actual goal of enterprise digitization to determine whether we need to collect equipment operation information to support equipment management and operation and maintenance system, or need to collect equipment start-stop, operation rate to support production quality management system and so on.

Then there is the sensing device layer, when we BL0170-PC0010-P know what data we need, we can choose different sensing devices according to the actual needs. For example, through sensors, cameras and other intelligent terminal data acquisition modules to collect signals, videos, pictures and other different data. When we collect the device data, we need to transfer it to the computer, cloud platform or edge computing device through the network. At present, the common enterprise communication network is 4G, industrial Ethernet or WIFI. However, considering the huge amount of data generated by the device layer every day, the cost, speed, coverage, life and so on of the network need to be measured by the enterprise. Among them, LoRa has become the choice of many enterprises because of its sensitivity (-148dbm), anti-interference ability, distance and cost.

In addition, in order to improve data quality, BL0170-PC0010-P after we obtain the original data, we usually carry out data cleaning, such as deleting duplicate information, correcting error information, identifying incomplete information and other methods to filter the data that does not meet the requirements. If we go further, we can also carry out data analysis and processing such as feature extraction and feature selection according to our needs, so that the data can be better used for us.

You can twist your back with a big step. Settle down, take the problem as the guide, start from data collection, and build a solid digital transformation base step by step to achieve progressive transformation.