#innominds

Implementation of Artificial Intelligence in industries has enabled smart manufacturing systems by integration with technologies such as cloud computing, big data analytics and connected devices. The transformative impact of Artificial Intelligence in different sectors such as healthcare, retail, finance, and transportation are undeniable. To unlock full potential of AI in asset-heavy manufacturing and production industries, Hybrid AI is the key.

Hybrid AI is the combination of non-symbolic AI such as deep learning and machine learning systems with symbolic AI such as the embedding of human intelligence. Hybrid AI systems have the potential to solve complex issues such as imprecision, uncertainty, and vagueness as it combines both data as well as knowledge to solve problems instead of learning everything from the data.

Classic applications of AI such as natural language processing and object recognition has a vast amount of training data available, but there is very less relevant sensors and other equipment data available in case of asset-heavy industry. For this, implementation of Hybrid AI with combination of data science and physics is the way forward to reduce the errors and improve efficiency.

Unlike the mathematical models used for consumer behavior analysis, industrial data needs to be validated by taking into account the environmental as well as physical conditions. To make industrial hybrid AI work properly, implement the essential tools and map the data from all relevant sources. Hybrid AI is best suited for industrial processes where machine learning models built on a mathematical theory framework is used on real-time data for predictions. This helps in insights driven decision-making process that optimizes production and creates value.

With right data and techniques, you can lower the operating costs, increase top-line revenue and profitability. Innominds’ has developed Numix, an advanced video analytics solution based on computer vision and deep learning. It provides wide range of solutions such as complete industrial surveillance, security, and anomaly detection in the busy production lines, to bring manufacturing efficiency. To know more about the secure and smart video analytics solution, get in touch with us today.

To know about A global auditing major improves support team’s productivity by 15% and reduces average time to resolution by 20%.

Download Case Study

OutSystems automatically offers basic CRUD methods for accessing the data, and you can develop your own aggregates. OutSystems displays a preview of the fetched data so you can see how modifications affect the outcomes right away. Custom SQL queries can also be used to retrieve data in OutSystems. As a result, the access pattern is extensive. This demonstrates that programming with OutSystems necessitates coding skills and is not all magic. If you are like me, who prefers to use aggregates as much as possible instead of the SQL widget, but, not able to do so because of some missing features like the ability to query with an IN clause, let me show you some hacks. Consider that we have an Employee table linked to a Department table as depicted in the following Entity Diagram:

The diagram below shows how a typical screen listing would look like:

Let us say we like to add a special filter to get the list of employees by Department Name. We may add another input box and bind to a variable, say – “Department Filter”.

Typically, we would modify the aggregate to accommodate the department filter like below:

This works fine when we want to filter by any single department name, like below:

However, if we want to modify the same aggregate to allow more than 1 department name (as we do in an SQL widget with comma-separated values using IN clause), there is no out-of-the-box way to do so.

So what is the solution?

Let’s hack it. Just use the Index function in the filter like below:

Just note that this may not work as expected in all the cases and may require some additional hacks. In case, the department name for example had similar words. For example, while searching for only the “Research” department, the result would also include “Research & Development” department values.

Conclusion:- 

The real advantage of low-code is that it allows more individuals to code because the learning curve isn't as steep. That isn't to say you don't need to know about software architecture; you do. Otherwise, maintaining and developing the systems will become extremely difficult. This means that businesses will develop more applications both internally and outside. As a result, rather than replacing existing solutions, there will be more of them, and applications will be developed by a larger set of developers. As a result, the demand for new functionality and APIs for other systems, such as existing backend solutions, would increase. Thanks for taking the time out to read this blog! Keep an eye on this space for a series of more interesting blog posts around OutSystems in the coming days. Note:- The blog is originally published on Innominds Website