U-NOVUS 2018: workshop
In mid-October, we held a workshop on Data Science as part of the U-NOVUS youth forum in Tomsk.
Tomsk, in principle, deservedly enjoys the fame of the city of scientists and students, after all 15 research institutes, 9 universities and several business incubators - this is serious. Therefore, we decided to invite both students and experts from various companies to participate.
')
We gave the case from life (read - from production), it was the task of advanced analytics at a petrochemical plant.
About how it was - under the cut.
The workshop lasted 3 days, it was just so much time for the teams to solve our tasks and show that the solution they created is what the industry will actually help, or it will just carry a number of useful mechanics that can be applied on the production of digital chemistry in the future.
It was necessary to create a working scenario in which the development and implementation of a proactive monitoring system of technological equipment that we use in production would be carried out.
At the same time, it was important to take into account that such equipment is logical to be divided into several types (by severity), so the approaches to their management and monitoring should not be the same, and a single script would not work here. It was also necessary to take into account that the enterprise uses the simplest visualization systems of already collected data, which can also be used. Plus we gave a number of factors to the load - the influence of the state of the equipment on product margins; the frequency of scheduled repairs; scenarios for installing additional stands in cases where the basic monitoring system already exists, and so on.
And we immediately prescribed a number of frameworks and restrictions that must be taken into account, otherwise it turns out that you made the decision, but you cannot apply it, because you forgot about one of these factors. This helps, because such a solution should work in live production, and there in the process a lot of different things can happen.
Among these factors were:
Required components of the system: a module that finds anomalies in the equipment (something heats up, but should not, something dangles, but similar behavior should hold), and a prediction module that can predict a similar situation based on the data already collected .
At the exit, I wanted to get a detailed description of the solution, which will, taking into account all these conditions, introduce a proactive monitoring system for the equipment. It was possible to include machine learning algorithms, any ready-made solutions and frameworks.
And quite ideally (and this is why the teams included people from the business) - to mention those business processes that will be affected by the implementation of such a system; you may even have to introduce new business processes to ensure the work of the solution.
We must pay tribute to the teams - they showed themselves perfectly. The teams were rather disparate; within the framework of one, students, programmers with data analysts, directors of directions, and directors of local companies could immediately work. And such a composition greatly influenced the solutions obtained as a result, we checked and immediately noted that someone had a strong emphasis on the architectural part, someone put interaction with users at the forefront, someone decided that the main thing is KPI planning and compliance. In general, you look at the solution - and immediately imagine who exactly was inventing it.
Our evaluation criteria were fairly simple. The main thing is the practical applicability of the solution in our enterprises. Almost everyone coped, of the 6 solutions presented to us, only 2 did not fit at all (although, in a sample of 6, this is a third). But there the thing was that the guys either failed to work out the solution itself, without going into details, or the solution was not suitable for the petrochemical industry. Alas, it also happens this way - and the solution itself seems to be a good one, it solves problems, maybe it even scales, but we don’t specifically apply at all, the stack is not the same. At all.
The remaining 4 decisions showed themselves perfectly, we decided that the guys understand exactly what they did and what they will do, so they will now participate in our projects.
Nikolay Ksenzik, Head of Digital Technologies Center in Tomsk, SIBUR IT.
Tomsk, in principle, deservedly enjoys the fame of the city of scientists and students, after all 15 research institutes, 9 universities and several business incubators - this is serious. Therefore, we decided to invite both students and experts from various companies to participate.
')
We gave the case from life (read - from production), it was the task of advanced analytics at a petrochemical plant.
About how it was - under the cut.
The workshop lasted 3 days, it was just so much time for the teams to solve our tasks and show that the solution they created is what the industry will actually help, or it will just carry a number of useful mechanics that can be applied on the production of digital chemistry in the future.
Task
It was necessary to create a working scenario in which the development and implementation of a proactive monitoring system of technological equipment that we use in production would be carried out.
At the same time, it was important to take into account that such equipment is logical to be divided into several types (by severity), so the approaches to their management and monitoring should not be the same, and a single script would not work here. It was also necessary to take into account that the enterprise uses the simplest visualization systems of already collected data, which can also be used. Plus we gave a number of factors to the load - the influence of the state of the equipment on product margins; the frequency of scheduled repairs; scenarios for installing additional stands in cases where the basic monitoring system already exists, and so on.
And we immediately prescribed a number of frameworks and restrictions that must be taken into account, otherwise it turns out that you made the decision, but you cannot apply it, because you forgot about one of these factors. This helps, because such a solution should work in live production, and there in the process a lot of different things can happen.
Among these factors were:
- Poor quality of the data collected.
- The presence in the team for the creation of such a monitoring system specialists with different expertise - programmers, production technologists, people familiar with the equipment, as well as experts in the field of mathematical modeling.
- The reluctance of staff to use the new system (well, otherwise how else).
- Complete lack of data for solving problems (some parameter is not considered, there is no sensor that would remove information, etc.).
- If there is any data, there may be difficulties with it - for example, not all of them are digitized (and you need to work with them), they are stored in different formats, some are not even accessible in a couple of clicks, and you have to go through a couple of laps approvals.
Required components of the system: a module that finds anomalies in the equipment (something heats up, but should not, something dangles, but similar behavior should hold), and a prediction module that can predict a similar situation based on the data already collected .
At the exit, I wanted to get a detailed description of the solution, which will, taking into account all these conditions, introduce a proactive monitoring system for the equipment. It was possible to include machine learning algorithms, any ready-made solutions and frameworks.
And quite ideally (and this is why the teams included people from the business) - to mention those business processes that will be affected by the implementation of such a system; you may even have to introduce new business processes to ensure the work of the solution.
Total
We must pay tribute to the teams - they showed themselves perfectly. The teams were rather disparate; within the framework of one, students, programmers with data analysts, directors of directions, and directors of local companies could immediately work. And such a composition greatly influenced the solutions obtained as a result, we checked and immediately noted that someone had a strong emphasis on the architectural part, someone put interaction with users at the forefront, someone decided that the main thing is KPI planning and compliance. In general, you look at the solution - and immediately imagine who exactly was inventing it.
Our evaluation criteria were fairly simple. The main thing is the practical applicability of the solution in our enterprises. Almost everyone coped, of the 6 solutions presented to us, only 2 did not fit at all (although, in a sample of 6, this is a third). But there the thing was that the guys either failed to work out the solution itself, without going into details, or the solution was not suitable for the petrochemical industry. Alas, it also happens this way - and the solution itself seems to be a good one, it solves problems, maybe it even scales, but we don’t specifically apply at all, the stack is not the same. At all.
The remaining 4 decisions showed themselves perfectly, we decided that the guys understand exactly what they did and what they will do, so they will now participate in our projects.
Nikolay Ksenzik, Head of Digital Technologies Center in Tomsk, SIBUR IT.
Source: https://habr.com/ru/post/432258/
All Articles