App settings:
AI in Product Creation
What does AI mean?
The term AI can be traced back to the mathematician Alan Turing, who in 1950 formulated the idea that a machine could have the same cognitive abilities as a human. Even though, the understanding of the term AI has continuously changed over the recent years. Because of the continuous changes in what technology is capable of as well as the assertiveness of various interest groups, the definition of AI is subject to an ongoing change. So, a unified definition of AI seems to be accompanied by various challenges. Most AI applications are very specialized in terms of their tasks. Such AI Applications are called weak AI.
Our understanding of AI in product creation
In product creation is determined which market output a company wants to offer to future markets. So, product creation is a key driver for product innovations, which have significant importance for Germany's economy. Since products are getting more complex and development cycles are getting shorter, the producing industry is faced with the challenge to increase the efficiency of their product creation processes. Leveraging the power of Artificial Intelligence is one way to master this challenge. Using AI offers companies multiple opportunities to improve their products, services, and production processes. Correctly used, AI also offers new opportunities for data analysis and evaluation and paves the way for the development of more innovative, tremendously improved products with a significantly increased perceived customer value.
AI-Tags
Computer Vision (CV)
Computer vision includes all processes for handling image and video data. In particular, CV includes activities for analyzing and understanding images taken by a camera.
Example of use:
A typical example is object recognition. Object recognition can identify objects like components by analyzing, localizing, and naming these objects in a picture.
Knowledge Discovery
The area of Knowledge Discovery includes not only knowledge discovery but also presentation activities. This includes, in particular, the discovery of knowledge in large datasets by using data mining methods as well as dealing with unstructured data for instance in the shape of knowledge graphs. Mainly, unsupervised learning is used in this case, which means training without prior labeling of the data.
Example of use:
A typical example is cluster analysis, which is the clustering of data regarding underlying similarity structures.
Natural Language Processing
The area of Natural Language Processing includes all methods for dealing with both spoken and written human language. This includes for example the analysis of requirement documents or customer reviews.
Example of use:
Examples in product creation context include the analysis of requirement documents or customer reviews..
Decision Support
The area of decision support deals with methods for decision support in complex environments.
Example of use:
This includes case-based decision-making. For this, analogies are used so that a similar solution can be applied to the current problem. In a nutshell, applying similar solutions to similar problems.
Modeling Languages
The area of modeling languages deals with formal models. In the context of product creation, this includes graphical visualizations of modeling languages like UML or SysML.
Example of use:
Examples of high significance are the transformation of platform-independent models into platform-specific models or the translation between different formalisms.
Signal Processing
The area of signal processing includes dealing with signal data. Signal processing is focused on all applications of time series as well as common sensor data, like vibration or temperature data. Supervised learning methods are used, meaning that the training data set is already labeled.
Example of use:
An example in terms of product creation is the analysis of machine signals for identifying anomalies and related defective behavior.
