Data are increasingly important in many industry sectors. Plants managers often do not realize the importance of having a huge amount of data available. The implementation of Key Performance Indicators (KPIs), based on these data, can be really helpful to give to the managers an overview of the real behaviour of a plant, in order to improve its efficiency and respect the local limits.
Python has become one of the most appreciated programming languages in data mining, as well as in decision tree algorithms development, due to its specific libraries for data handling. Combination of different sensors data, plant managers and operators feedbacks and a deep industrial process knowledge, is a great set of skills to get KPIs tailored on plants characteristics and management needs.
Operational use cases of KPIs and decision trees algorithms developed in Python are presented, as demonstration of how different engineering skills mixed with the power of Python programming could really boost the work quality and the customer satisfaction.
Introduction to Microprocesso programming and interfacing.pptx
KPIs implementation and decision tree algorithms as support tools in wastewater treatment plants management
1. KPIs implementation and decision tree algorithms
as support tools in wastewater treatment plants
management
Giuseppe Antonello, ETC Sustainable Solutions (Trento)
Claudio Modena, Lorenzo Rizzoli, Angelo Cantatore,
Marco Vian, Elisabetta Fresta, Francesca Bellamoli,
Daniele Renzi, Francesco Gelmini, Debora Agarossi
Trento, 30th
August 2018
2. 30th August 2018 EuroSciPy 2018, Trento 2
Business of data
Who are the biggest masters in the
business of the data?
8. 30th August 2018 EuroSciPy 2018, Trento 8
DB architecture
DB_instantaneous
DB_daily
DB_raw / csv
9. 30th August 2018 EuroSciPy 2018, Trento 9
KPI
Key Performance Indicator
KPI is a type of performance measurement. KPIs evaluate the success of
an organization or of a particular activity in which it engages. (Wikipedia)
10. 30th August 2018 EuroSciPy 2018, Trento 10
iSg: macro-indicator of the behaviour of the biological sector
of a plant. With the aggregation of power and oxygen values,
it indicates if a plant works in a pre-validated operating range.
Some KPI in our use case
iSO: index of the operative stability. It indicates the “load
state” of a plant.
iQ: this indicator refers to the specifc consumption / m³ of
treated wastewater. It “talks” about low and high load of a
plant.
iC: specifc consumption, done by the correlation between the
avg ammonia nitrogen and ammonia concentration. It points
out a probable malfunction of the ammonia probe.
IE: energy indicator. It measures how much energy is wasted,
dosing more oxygen than necessary.
11. 30th August 2018 EuroSciPy 2018, Trento 11
Class defnition of KPIs (1)
12. 30th August 2018 EuroSciPy 2018, Trento 12
Class defnition of KPIs (2)
16. 30th August 2018 EuroSciPy 2018, Trento 16
Conclusions
Python really helps and eases KPIs implementation
KPIs and decision trees are a good tool to help customers
Customers need and love Dashboards, if well designed...
Dashboards are well designed if we are able to think as a
customer, not as a nerd...
A good communication is essential to valuing the (night)
hours spent in front of a monitor
Python is an appreciated toy also in industry!
17. 30th August 2018 EuroSciPy 2018, Trento 17
Thank you
We’re hiring!
Contacts: giuseppe.antonello@etc-eng.it
ETC Sustainable Solutions, Trento
http://www.etc-eng.it