In this keynote presentation at the ASQF-NRW Testingday, Rik Marselis (Principal quality consultant at Sogeti) explains how Quality Engineering can be implemented in high-performance IT delivery teams by applying Artificial Intelligence, Machine Learning and other ways of automation to ultimately achieve quality forecasting which enables to solve problems before people notice any problem!
This presentation is based on the books "Testing in the digital age - AI makes the difference" and "Quality for DevOps teams", both from the TMAP body of knowledge. Rik is co-author of both books.
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
Quality engineering in the digital age... Why? How? (ASQF Keynote by Rik Marselis)
1. Quality Engineering
in the digital age…
Why? How?
Rik Marselis
ASQF NRW
15 April 2021
TMAP: the body of knowledge for
quality engineering in IT delivery
You can win a copy
of this book!
(the ePub version)
19. AI based analytics and forecasting
Machine intelligence is based on:
• Descriptive analytics
• Predictive analytics
• Prescriptive analytics
A major benefit is…
Forecasting
23. 23
Machine Learning (as the most common way of implementing AI)
▪ Machine Learning is one of the ways to
achieve artificial intelligence.
▪ It contains different algorithms – each with
its own strengths and weaknesses.
▪ These algorithms are often grouped into
three categories:
• Supervised learning
• Unsupervised machine learning
• Reinforcement learning
See the article of
Vijay Raghani
on SogetiLabs.com
24. Machine Learning categories
▪ Supervised learning is the machine-learning task of learning a
function that maps an input to an output based on example input-
output pairs. It infers a function from labeled training data consisting
of a set of training examples.
▪ Unsupervised learning is the machine-learning task of inferring a
function to describe hidden structure from "unlabeled" data (a
classification or categorization is not included in the observations).
▪ Reinforcement learning is a variety of machine learning that
determines various options to find the option that maximizes some
notion of cumulative reward. Reinforcement learning differs from
standard supervised learning in that correct input/output pairs are
never presented, nor sub-optimal actions explicitly corrected. Instead
the focus is on performance.
27. This is also called
“Feature
engineering”
(the variables are
called features)
Machine learning is a complex process
which is not fully automated
Select
data
Preprocess
data
Transform
data
Train
model
Preparation of the data Actual
machine
learning
Largely manual!! Mainly automatic
29. Multi-model predictions
▪ Predictive analytics uses multiple models
▪ The available data from test execution and
live monitoring is fed into multiple models.
▪ Each model calculates the expected
evolution of quality.
▪ The results are shown in a “plume”. Today
▪ The digital quality engineer determines, based
on the relevant parameters, which is the most
likely future situation. If the quality is at risk a
“code red” can be issued.
▪ Based on this forecast the team can decide
what actions need to be taken to ensure
quality.
Near future
▪ Prescriptive analytics determine the most
likely situation and if quality is at risk
immediately triggers corrective action.
45. Download an information package using this link:
https://fts.capgemini.com/pubpwd/365156930079/ASQF_NRW_keynote_QualityEngineeringInTheDigitalAge_RikMarselis.zip
username: ajdfdhfsj Password: SCZ7WhRkAJ
It contains: preview-versions of both books, the article on the elevator
that looks into the future and info on the TMAP certifications scheme.
And visit the renewed
TMAP body of knowledge
on www.tmap.net
2018
2020
You can buy a book on
www.ict-books.com