How chemical companies can fight the tough battle of data governance

Information technology has now penetrated into various industries, and the chemical industry is no exception.
Various management systems such as ERP, MES, LIMS, equipment monitoring systems, safety information systems, intelligent inspection systems, personnel positioning systems, and OA systems have been have been applied on a large scale
.
But with the increasing number of system construction, some new problems have emerged
.
These systems originally built around the business of a specific department of the enterprise, due to the lack of unified planning, the characteristic data used to describe a business or thing is incomplete, and it is difficult to achieve the effect of comprehensive data utilization
.

In addition, when a department is building its own information management system, in order to avoid the communication barriers between departments, some businesses may be over-extended and developed within the system
.

From the perspective of its own development system, this is not a problem, but from the perspective of ordinary users, when faced with several or even a dozen large and small systems of the enterprise at the same time, over-extending development means over-crossing.
Repeatedly, users need to do some similar or even overlapping work in different systems, which not only reduces work efficiency, but also lowers users’ overall perception of enterprise informatization, and even more serious is the statistics of the same type of data collected by different systems The results may be quite different.
When facing the overall assessment of the company, each department holds a set of data.
The public says that the public is right, and the mother says that the mother is right.
Instead, the construction system has become the initiator of the "departmental wall"
.

If a system is only for the purpose of moving the original offline management and control mode to the online, in fact, it does not improve much efficiency for ordinary users, and sometimes it is not as convenient as offline management
.
The original intention of the business-oriented system design is not to consider humanized management, but to maintain the normal management order, strengthen process monitoring, and more importantly, be able to use the data from statistical analysis as the basis and support for management decisions, so the construction of the system If the needs of data sharing cannot be met, the traceability of data cannot be realized, and the comprehensive utilization value of data cannot be realized, the fundamental purpose of enterprise management informatization construction cannot be achieved
.

So, how can this fundamental purpose be achieved?

The key to solving these problems must be to effectively integrate resources in the system, so that various management services can flow more smoothly within the system and between systems, so that data can "speak", this governance process can be called Data governance
.

Specifically, the work should be carried out from the following aspects:

First of all, we must correctly understand the complexity and difficulty of this work.
We must consider the overall development level of the company and use global thinking.
We must mobilize all available resources for implementation.
Therefore, the data governance work must be led by the company.
This is the The prerequisites for driving data governance efforts
.
Also, be mentally prepared for a long battle, because it doesn't happen overnight
.

Under the leadership of the leader of the enterprise, a special data governance project team can be established, which is responsible for the following tasks:

One is to verify the availability of existing systems
.
The systems that cannot be used effectively and are quite different from the actual business need to be renovated and improved, and the systems that do not have the conditions for rectification and are too difficult to rectify should be discarded or rebuilt
.

The second is to sort out whether there are overlapping business functions in the existing system
.
For systems with repeated and overlapping functions, the most preferred ones should be evaluated, and the second preferred ones should be discarded.
At the same time, attention should be paid to the data connection between systems to ensure that the normal operation of each system will not be affected by discarding some functions
.

The third is to evaluate whether the data obtained by each management business of the existing system is complete
.
Businesses with poor data integrity should supplement and improve system functions and forms
.
There is a completeness criterion here - to judge whether the business can obtain a set of data to accurately describe the content, progress and implementation effect of the business
.

The fourth is to establish a unified integrated management and control platform to integrate all existing systems into the same system platform for management, which is not only convenient for users to operate and use, but also helps to break through the "information island" formed by the original system for a long time
.

The above solution is only the first stage of work based on the existing management system of the enterprise
.

In the second stage, starting from the level of corporate strategic planning and target planning, it is necessary to study which key indicator data should be monitored in the process of achieving these strategies and goals.
Do these data exist in the existing system? If not, can it be monitored through data consolidated from multiple systems? This process needs to be decomposed step by step from the "top of the pyramid" of enterprise management to the "base of the pyramid", and the parts that need to be improved in the function of the system are reversely traced from the monitoring indicators, so as to maximize the potential value of the system
.

For example, chemical companies want to achieve zero safety accidents, zero environmental pollution, 5% reduction in annual energy consumption, and 10% reduction in annual carbon emissions.
In order to achieve this goal, what process monitoring parameters are needed to measure? In which systems are these process monitoring numbers distributed? How authentic and reliable? Is there data "noise"? If direct access is not available, what are the alternative monitoring metrics? What are the insurmountable red lines of these monitoring indicators? If data cannot be obtained from all systems, should a new system be built or an existing system with similar functions should be added? The process of answering these questions is the governance process of the second phase of data
.

Of course, data governance still has a higher stage of development, such as using data to train artificial intelligence algorithms and allowing the algorithms to directly participate in management decision-making.
For chemical companies in the current period, there is still a long way to go.
to think about it
.

The difficulties faced by chemical companies in the data governance process may be far more than imagined, such as the difficulty of system integration developed in different periods, the common problem of poor system life cycle management, whether to rectify the system or start a new set of development.
.
.
these Issues need to be further weighed in the implementation process
.

There are also some non-technical problems, such as the problem of user usage habits.
There may be many unreasonable factors in the use of a certain system, but a set of coping strategies has been explored, and I am used to dealing with problems in this way.
How to persuade them to make changes ? For another example, who will accommodate whom during the integration process of the systems used by different departments? There is no single standard answer to these questions
.

In general, data governance is an inevitable path for enterprises to develop to a certain stage under the background of the continuous increase of various management systems.
It is also a work worthy of every enterprise's attention, because data is very important for modern enterprises.
It has increasingly become a valuable intangible asset, which determines the future sustainable development of the enterprise
.