Shulou(Shulou.com)06/03 Report--

Background: large manufacturing equipment, which needs to be shipped in batches and installed by the customer on site (the installation process lasts more than 15 days)

Problem description: the user feedback a question. The basic logic is that if the "whether to update shipping" information is checked in the intelligent system, the system will update the CRM system automatically, and the site installation service team will automatically receive the corresponding information.

For logistics clerks who are not very clear about the master plan, this information is difficult to maintain timely and accurate. It will bring deviation to the operation of the business, of course, it will also affect the accuracy of "order execution Kanban".

Point of view of this article: today I would like to focus on the limitations of ERP systems, MIS systems, and software design thinking. This kind of design thinking makes many ERP implementation consultants form a fixed problem-solving mode of thinking. What I want to say is that to break this kind of thinking, this kind of thinking is very efficient in repairing the main line of business flow of information technology, following a very rigorous logic, but when building a path, if you want to achieve this rigorous logic, you have to pay a higher cost, users will find it very troublesome to use it, and sometimes they will abandon it, and users will say, "I use Excel management, which is much more simple and effective than this."

Voiceover: have you ever encountered this confusion?

First briefly describe this mode of thinking, we know that similar to SAP,Oracle EBS, basically through the background management end according to business requirements, through configuration to achieve the corresponding business management requirements. For example, SAP configures the business of each module through the transaction code spro. The master data has a lot of attributes, and different attributes will implement different businesses. That is, if you want to achieve A business, then you need to configure B, if you configure B, then you will achieve A business.

Why does this rule no longer apply when it comes to building paths? In fact, for the more advanced (Hign Level) business, the higher the level of abstraction, that is, the stronger the regularity, and to the business implementation end, there will be more variables, become less rigorous logic, and this is often the real world. For example, the logistics is scheduled to leave at two o'clock this afternoon, but it is delayed due to traffic jams, rain and snow or haze. It was planned to finish the plan tomorrow afternoon, but an employee asked for leave in an emergency. Of course, the actual business variables are far more than these two examples.

You would say that these are just special circumstances and will not happen very often. That's right! It's right here. We can't let users maintain data for this 5%. Unless the business value is so great that it has to be maintained. We actually have to accept tolerances within the allowable range. Choose the appropriate information solution and give up moderately. This is one of the points. For example, when we apply for work in the workshop, we adopt this idea. We do not need to accurately ask workers to report for work, we only need to estimate 25% of the total working hours, 50% and 75% of the manufacturing progress is relatively accurate. 100% is accurate and easy to do.

The second point is that all events actually operate according to probability, and the probability will change with the occurrence of events. According to Bayesian theorem, prior probability-new information-Bayesian theorem-posterior probability

Open Baidu APP to see more high-definition pictures

Through the accumulation of events, we can get closer and closer to the truth, and over time, we will get a data model with approximate normal distribution, so that, through hypothesis testing, through the operation of the data close to the truth, the probability of making the right choice will be very high, and the system will know more and more about the business. This is the application of machine learning algorithm. This area is still learning, I believe this is the solution to the problem of the path, but also the trend.

In view of the above problems, and similar detailed business branch scenarios, we can take two steps. The first step is the viewpoint one: the appropriate cost, the greatest benefit, and the second step is for the computer to complete the learning process, thus further simplifying and even replacing the user's operation!

Remarks: the core management ideas of ERP are mainly embodied in the following three aspects:

First, embody the idea of managing the resources of the whole supply chain.

Second, embody the ideas of lean production, agile manufacturing and synchronous engineering.

Third, it embodies the idea of pre-planning, in-process control and post-feedback analysis.

The hallmarks of a successful ERP application are:

First, the operation of the system is integrated, and the operation of the software spans many departments.

Second, business processes are rationalized, and business departments at all levels are reconstructed according to the fully optimized process.

Third, performance monitoring is dynamic, and the performance system can provide immediate feedback in order to correct the problems existing in management.

Fourth, continuous management improvement, enterprises to establish a mechanism for continuous self-evaluation and continuous improvement of management.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.