agilePHM – a new open source product for rapid prototyping of PHM analytics


We are launching a new product called agilePHM

In Industrial IoT, I have been working with PHM  (Prognostics and Health Management) for a while and it is a well known discipline

Prognostics and Health management(PHM)  is an engineering discipline focused on predicting the time at which a system or a component will no longer perform its intended function. This lack of performance is most often a failure beyond which the system can no longer be used to meet desired performance. The predicted time then becomes the remaining useful life (RUL), which is an important concept in decision making for contingency mitigation. Prognostics predicts the future performance of a component by assessing the extent of deviation or degradation of a system from its expected normal operating conditions.The science of prognostics is based on the analysis of failure modes, detection of early signs of wear and aging, and fault conditions. An effective prognostics solution is implemented when there is sound knowledge of the failure mechanisms that are likely to cause the degradations leading to eventual failures in the system. (source wikiedia)


PHM applies to a range of domains like defence, shipping, industrial applications etc

We are developing a product for rapid prototyping of the analytics component of PHM

agiilePHM is designed for for rapid prototyping of PHM applications (implemented on standard hardware)

The need for the product

The idea of agilePHM arose due to a few observations

a)  There is a need for rapid prototyping of new ideas(from an analytics standpoint) as an exclusive function: Industrial IoT is a very new space and evolving. Ideas from different domains cross-pollinate and there is a need to quickly test out concepts(either products or processes)

b)   Data Science skills shortage: Data science skills are expensive and are often focused on industries like Banking(in contrast to Industrial IoT). So, think of agilePHM like ‘Data Scientist in a  box’ for the Industrial IoT space

c)  larger products have a much heavier footprint: Our customer is someone who wants to rapidly prototype the model (without knowing the algorithms in detail). Larger products have a much heavier footprint. Many seem like installing ERP in the old days! They perform the function of rapid prototyping as a small component (as opposed to exclusive emphasis on it)

d)  Flexibility: The approach complements existing approaches like Physics based modelling

e)  Why open source ..  Our main strength lies in IoT analytics (Ajit Jaokar teaches a course on Data Science for Internet of Things at the University of Oxford). However, the problem we address is complex because there are many processes (and many machines!) to abstract algorithmically. This needs some form of open source.


agilePHM has three components

1) Digital Twin

2) Rapid prototyping

3) Workflow – Process engineering

agilePHM will have the following deployment models
On premise with support

Open source

Kaggle like contest community engagement

It also allows students in our course to gain real life / practical experience


a)   If you are a company interested in working with us, please email me on ajit.jaokar at

b) If you are interested in gaining real experience in AI .. you can work on the product with companies as part of our course. Please contact info at to know more