Automation and / or result?

Introduction

In the 1990s, the word “computerization” was very popular. At times, colossal funds were affected for the purchase of computers - even without an understanding of its role in increasing the efficiency of labor. The low literacy of most businessmen in the field of automation, often encountered in those years, can now cause us to smile, but I recommend to look around well and find out: what do your acquaintances who have their own business or who are in top positions know managers on this topic? What has changed?

I feel that I can absolutely unequivocally state that the literacy of most businessmen in matters of automation has moved, but is still far from the point where people will understand well what exactly they are paying for. Moreover, those who sell automated control systems, rarely themselves, for what they get the money. Low effectiveness of the implementation of the species with the naked eye: BI systems that are designed to provide tools to the end user, but in fact require the maintenance of a staff of the most powerful programmers and analysts, costly ERP, where behind the beautiful "facade" instead of the promised know-how, banal arithmetic is found, and performance monitoring systems that give incorrect tasks to performers. The cost of the error in our time is amazing: the incorrectly performed tasks can “eat” not that millions, but billions of rubles!

The specificity of my work is such that a lot of express audits have to be conducted as a result of the implementation of information systems. Only in the last 2 weeks I have been able to visit 4 enterprises that have demonstrated: an inefficient BI system, an incorrect WMS, just an idle MES, and before the heap there was a complete misunderstanding of the result. The most surprising thing was that the customer at the same time easily agreed to “write TK” for the performer, not understanding anything in the subject area, and the performer accepted the poorly structured text, to which the term “stream of consciousness” is more than “document ". Even more surprisingly, customers emphasized the following "positive" characteristics of implementation companies:
1) "Saved a lot of time, abandoning the formalization of processes"
2) “We agreed to carry out all the modifications directly in the mode of trial operation, without fixing them in the documents so as not to increase the project budget”
3) “True professionals! Even without a survey they told me what the settings would be in the system! ”
Now I propose to consider in turn the cases that I recently had to face.

Case 1: Ineffective BI

The company acquired a very expensive product when it learned that users will be able to independently develop all the reports they need. The sales manager colorfully presented an excellent concept, when the CEO, using his favorite tools, builds the most complex reports, predicts sales, receives notifications about the least efficient employees, and much, much more. There were only two obstacles between the reference point and the director’s dream: to conclude an agreement and pay for the product and the work. The surprise came a little later when consultants came to the head to get information from him about the reports he wanted to see. When he reminded them about “any report with their own hands,” it was gently corrected: you can make a report with your own hands, but before that you need to adjust the data model, and if you don’t have any objects in this model, use them in the reports will be impossible.
')
The head quickly made a decision: he convened all the top managers and the analytical department, and brought them directly to the consultants, ordering them to come up with the maximum possible number of objects for analysis. Having thought well for a few days, they presented their work to the consultants, for which they received another soft amendment: it is impossible to construct a query from objects that are not related to each other, so now all the possible views for analyzing the data are required.

I will not long enumerate the misadventures of management and analysts, let me just say that at the end they returned to what they were struggling with: in order to make a competent model, it is first necessary to develop a reporting system. The director has already given up on his dreams of artificial intelligence, which makes it possible to build reports from “I don’t know what”, and gave the go-ahead to the work of consultants to develop key performance indicators and reports that will allow evaluating the work of the departments. However, consultants performed their work in the same “highly efficient” mode: they simply asked key users what they wanted to see, and structured this gigantic amount of information a little.

Result: a server with fantastic characteristics, on which useless cubes are counted, in order to be able to look at the charts in the web interface. The manual was played by the query designer, but now uses the integrated reporting system of its KIS.

Conclusion: any tool requires competent application. To swing a hammer in different directions, dispersing mosquitoes is a completely ineffective occupation. However, in any project there is an investor, and some project managers believe that the project can be considered successful when the investor gets what he wants. Immediate investor satisfaction with the "toys" as a result came down to frustration, when the tool was absolutely useless. Needless to say, the customer no longer wants to work with this supplier.

Case 2: Incorrectly working WMS

In a small review article on WMS, I mentioned what basic functions are in such systems. Now imagine a warehouse where WMS is installed, but having a very reduced functionality, and yes even giving out incorrect instructions to performers. Moreover, there are problems not only in the functional part, but also in the implementation methodology.

The problems begin with the warehouse topology: the buffer places are not reflected in the system in any way, but are controlled by the warehouse workers. At the same time, there is no regulation in which even general information on how to exercise this control would appear. Huge cells are deployed on 12-15 pallets, and the label that is glued to the pallet has a very small size. This means that when performing a task, an employee is forced to look for a pallet with a specific number, reading each label, and reducing its productivity in short operations by 3-4 times. Further - more: when placed on the shelves, the system does not recommend exactly what place is required to place the cargo. The staff does it at their discretion, and his discretion is “came a pallet - we place a pallet, a box has arrived - we place a box”. At the same time, no one even thinks that some goods come in whole pallets, but leave slowly and sadly, so this pallet will last for 3-4 months. What is the point for such an unpopular product to occupy the whole pallet place if you can replenish the shelves?

I will not list them in total, because as a result of the two-hour run through the warehouse, a solid report has turned out, and most importantly, specific recommendations on increasing productivity. What is surprising: the contractor who introduced the product did not even write the technical task. He was guided by the wishes of the customer, and most of the functionality was redesigned already during the operation of the system. The customer considered this advantage: the contractor’s staff were ready to carry out improvements right according to his words! However, the reality turned out to be much more sad: not a single document indicated these improvements are fixed, and the guarantee from the supplier extends only to the “boxed” functionality, the description of which the customer received upon payment of the license.

It is also surprising that the contractor was in full confidence that he worked perfectly well: his staff took part in far more than one dozen projects, and he worked out - in his opinion - “best practices”. The only problem is that the "practices" are not always the "best." If what is repeated from project to project, and does not cause overt problems, is called “best practices”, then the consumption of arsenic in small amounts is also “best practice”. When it comes to the fact that it is harmful, it will be too late for the one who invented this, and for those who have taken such a “useful” initiative.

Summary: now we need to write a full-fledged TZ for setting up the acquired system, and if it is not possible to set up, you will have to change the product by spending money again. But before the development of TK, one should invest in the development of a rational technology of cargo processing in order to have information on the specific benefits that will be obtained after automation. WMS is, again, only a tool that allows you to fix the process, and to ensure its implementation, issuing relevant tasks to the performers. The lack of substantive knowledge of the customer and the contractor led to the fact that "one said wrong, and the other misunderstood."

Case 3: Non-MES

In this case, I would not use the term MES, since we are talking about a system that does not plan equipment loading and distribution along the lines, but ensures - just like WMS - compliance with the technological process, but not in stock, but in production sites. However, in the absence of the best terminology, I will attribute this system to the class MES.

The customer has a small production, which implemented quite complex processes with the transfer of materials between different technological areas, there is a variety of processing and the formation of semi-finished products, the mandatory use of the weight complex, and many employees, each of which is unique in its own way. Naturally, no one wants to be dependent on whether Ivan Vasilyevich comes to work tomorrow, which is why the management decided to implement a system that would allow it to adapt more quickly to the workplace and reduce the staff qualification requirements. Plus, I want to track stocks of materials in a variety of industrial warehouses.

In general, they contacted the local system integrator, who coordinated the TZ, implemented the specified processes, and “rolled out” their product to the customer. And then - as in the best comedies: the customer looks at the product, and does not understand: it seems that everything they wanted was realized, and what we have at the output is not clear. So, without having understood, and accepted a product, having signed even the press release about successful use. Then they sat in a circle in front of the computer, and began to learn. Not understood. Again they tried - again they did not understand.

Let me just say that you can not laugh like this every day, week, or even half a year. The contractor implemented the customer's wishes in the form of an electronic document management system, where instead of stationary and mobile workstations, sensors on the lines and integration with controllers, a banal “paper” scheme was implemented: a Production Plan document is being created, which is closed with documents like “Finished goods release” and “ Production of semi-finished products. Each operation was implemented by a separate document, the interface and functionality of which did not imply its use in the workplace. Instead of a system that should issue tasks, they wrote a system that performs a simple accounting function.

findings

The cases described are only the tip of the iceberg. I can cite many examples where the customer literally rested his arms and legs, defending his position, so that after a few months he would gloomily say: “It seems I understood.” And someone does not even recognize, defending their position and losing competitiveness just because they lack the character to admit they were wrong. Yet the human factor is a powerful force that can be used both for good and for harm. But the customer is also a customer in Africa. He has money, he pays them, and therefore has the right not to know, being ready to pay for this ignorance. More surprising suppliers who look like jinn with a dirty trick: every desire must be thought out in detail, recorded, and signed. However, if we look at the European experience, then such an approach is rarely used: professional solution providers develop their competencies and are ready to offer a lot of us cases for their products, as well as develop an individual solution with a full-scale feasibility study. No one frees a business from adhering to the basic principles of ethics, and if you “called yourself a load,” then “get into the body,” and gain knowledge about the area in which you work. Do you know what profession is now in great shortage? It is analysts. Programmers - in bulk, managers - in the same way, but competent analysts - in the afternoon with fire to search. From this, the effectiveness of projects suffers: introducing products literally with eyes closed, relying only on luck is too expensive, for which — as a result — everyone pays: customer — money, supplier — reputation.