Our mission is to accelerate your business by creating custom predictive software tailored to yield maximum value for your business. To this end, we use both machine learning technology and mathematical models; and stand out when process-based models can complement data.
The creation of predictive software involves the set-up of a digital twin which comprises the virtual representation (modelling) of a business process, physical system or operational process. This model (or digital twin) can then be used to predict, and optimize ongoing real-world processes.
Throughout the entire process of setting up a digital twin, we can guide you from proof of concept to production.
To this end, we keep in mind the TDS process model which entails several iterative stages.
PropheSea can be of assistence during each stage of the process and provide guidance to:
- identify and fine-tune use cases
- evaluate if and how predictive solutions can accelerate your business
- build custom predictive solutions based on data and process knowledge
- deployment of tailored predictive solutions
Development of digital twins can be approached in various ways, depending on the nature of the system being modeled and the available resources. Two prominent methodologies for developing digital twins comprise (1) mathematical modelling combined with computational techniques and (2) leveraging machine learning technology driven by data.
Mathematical models are prominent when the system's behavior is well described and can be accurately captured by scientific principles; whereas machine learning stands out where data is abundantly available and the process is difficult to describe within a mathematical model.
At PropheSea, we develop digital twins using both machine learning technology and mathematical models, and particularly shine when facing problems where data can complement process knowledge (and vice versa). This allows to create digital twins which combine mathematical models with machine learning technology, providing various advantageous:
An illustrative industry example on how PropheSea combined process-based models with machine learning technology is provided here.
Check here for more details on how machine learning technology can improve mathematical modelling.
We aim to build trustworthy predictive models. Thereto, we seek transparency and explainability in our models; and pursue the mitigation of any unfair bias. We are committed to provide secure solutions which protect personal information.
In order to build a trustworthy relationship, we align ourselves as a partner for our clients and stakeholders while developing your innovation. Similarly, we set the bar high in terms of code of ethics.
