Blockchain-Orchestrated Intelligent Water Treatment Plant Profiling Framework to Enhance Human Life Expectancy

Dhruv Sarju Thakkar, Aneri Thakker, Rajesh Gupta, Nilesh Kumar Jadav, Sudeep Tanwar, Giovanni Pau, Gulshan Sharma, Fayez Alqahtani, Amr Tolba

Research output: Contribution to journalArticlepeer-review


Water quality degradation has turned out to be of crucial importance due to various factors over the past decade. Pollution, climate change, and population growth are the factors that affect water quality. Contaminations such as microorganisms, heavy metals, and excessive nitrogen and phosphorous disrupt water pH levels, posing significant health risks. Despite the innovation in the Internet of Things(IoT), allowing balancing the pH by adding chlorine and fluoride after the disinfection step, several security issues(e.g., distributed denial of service, data manipulation, and session hijacking) manoeuvre the operational performance of the water treatment plants. This causes people to consume polluted water, which has many adverse effects on human health and reduces life expectancy. To address this critical concern, we propose a novel approach integrating artificial intelligence(AI) and blockchain technology into water treatment plant management. Our methodology utilizes a standard water quality dataset, which has features such as pH and total hardness, which is used for binary classification, indicating water as potable or not potable. We employ various AI classifiers such as stochastic gradient descent classifier (SGDC), decision tree (DT), Naive Bayes (NB), K nearest neighbours (KNN), and logistic regression (LR). Furthermore, an InterPlanetary File System(IPFS)-based public blockchain is integrated to resist the data manipulation attack, where the potable water sample is securely stored in the blockchain's immutable ledger. The proposed model is evaluated using various performance metrics such as confusion matrix analysis, learning curve assessment, training accuracy, and blockchain scalability. Notably, the DT model emerges as the best-performing classifier with an accuracy of 99.41% and scalability of 35 with 120 data transactions.

Original languageEnglish
Pages (from-to)49151-49166
Number of pages16
JournalIEEE Access
Publication statusPublished - 2024


  • Artificial intelligence
  • blockchain
  • Internet of Things (IoT)
  • IPFS
  • water profiling
  • water treatment plants

ASJC Scopus subject areas

  • General Computer Science
  • General Materials Science
  • General Engineering


Dive into the research topics of 'Blockchain-Orchestrated Intelligent Water Treatment Plant Profiling Framework to Enhance Human Life Expectancy'. Together they form a unique fingerprint.

Cite this