Skip to content
DraconiticResearch
Search Sources API
182,544 studies31 sourcesUpdated 59 min ago
Peer-reviewedCC BY 4.0ACS Omega

Data Characterization and Multi-Plant Data Integration Strategies for Artificial Neural Network Modeling in Palm Oil Refining

  • Nur Syuhada Mohd Ali1
  • Tengku Zulaikha Malim-Busu2
  • Shahrum Shah Abdullah2
  • Mohd Fauzi Othman3
  • Zainal Ahmad1
  • Khairiyah Mohd-Yusof4
4 affiliations
  1. 1Universiti Sains Malaysia , , , ,
  2. 2Universiti Teknologi Malaysia , , , ,
  3. 3Tokyo City University , ,
  4. 4Purdue University School of Engineering Education , , ,
Open at sourcePDFDOICite

Abstract

Real-time quality monitoring in palm oil refining is often hindered by laboratory delays, which can be mitigated through the development of Artificial Neural Network (ANN) soft sensors. However, the effectiveness of these models is constrained by the characteristics of available process data. To address this, this study establishes a quantitative framework for assessing data set suitability by analyzing industrial data from three refineries using coefficient of variation (CV) statistics and self-organizing map (SOM) visualizations, with a feedforward neural network (FFN) used as a baseline model. Results show that process variables with CV exceeding 50% consistently yield better FFN performance than those below 50%. An integration strategy between different refineries was introduced to overcome the limitation of single-plant data and improve model performance to above 80% prediction accuracy for FFA content and Red Lovibond color. To test whether this improvement resulted from model capacity rather than data variability. The results further showed that data variability significantly affects model performance rather than model architecture. These findings establish CV as a practical criterion for data set suitability in ANN modeling for palm oil refining, offering an approach other chemical process industries facing similarly low-variation data may adapt.

Related

  1. From Fiber Architecture to Digital Twins: Characterization, Orientation Modeling, and Artificial Intelligence in Short Fiber-Reinforced Polymer Composites
    Ion-Dragoș Uțu, Diana Uțu
    PolymersPeer-reviewedCC BY 4.0Full textMaterials & polymersCrossrefEurope PMC
  2. Sustainable Bio-Based Polymer Adsorbents for Water Remediation: A Systematic Review of Functionalization Strategies, Performance Enhancement, and Circular Economy Integration
    Segundo Rojas-Flores, Moisés Gallozzo-Cardenas, Santiago M. Benites +2
    PolymersPeer-reviewedCC BYFull textMaterials & polymersEurope PMC
  3. Small-Strain Tensile Characterization and Finite Element Calibration of Sodium Alginate-Based Biomimetic Artificial Muscle Strips
    Xuan Wen, Lan Wang, Yan Xu +3
    ACS omegaPeer-reviewedCC BY-NC-NDFull textGeneral chemistryEurope PMC
  4. Sustainable strategies to improve indoor comfort and reduce energy consumption in winter: Integration of fly ash and rice husk ash in PCM concrete
    Mingming Zhang, Qingling Wu, Tingting Lu +1
    PLOS OnePeer-reviewedCC BY 4.0Full textMaterials & polymersPLOS
  5. Artificial Intelligence- and Machine Learning-Driven Strategies for Catalyst Design and Sustainable Chemical Processes
    Amra Bratovčić, Vesna Tomašić
    ProcessesPeer-reviewedCC BY8 citedMaterials & polymersOpenAlex

Cite

BibTeXRIS
BibTeX
@article{ali2026data,
  title = {{Data Characterization and Multi-Plant Data Integration Strategies for Artificial Neural Network Modeling in Palm Oil Refining}},
  author = {Nur Syuhada Mohd Ali and Tengku Zulaikha Malim-Busu and Shahrum Shah Abdullah and Mohd Fauzi Othman and Zainal Ahmad and Khairiyah Mohd-Yusof},
  journal = {ACS Omega},
  year = {2026},
  month = nov,
  doi = {10.1021/acsomega.6c04412},
  url = {https://pubs.acs.org/acsodf/article/doi/10.1021/acsomega.6c04412/5462348/Data-Characterization-and-Multi-Plant-Data},
  note = {CC BY 4.0},
}
RIS
TY  - JOUR
TI  - Data Characterization and Multi-Plant Data Integration Strategies for Artificial Neural Network Modeling in Palm Oil Refining
AU  - Nur Syuhada Mohd Ali
AU  - Tengku Zulaikha Malim-Busu
AU  - Shahrum Shah Abdullah
AU  - Mohd Fauzi Othman
AU  - Zainal Ahmad
AU  - Khairiyah Mohd-Yusof
PY  - 2026
DA  - 2026/11/02
JO  - ACS Omega
DO  - 10.1021/acsomega.6c04412
UR  - https://pubs.acs.org/acsodf/article/doi/10.1021/acsomega.6c04412/5462348/Data-Characterization-and-Multi-Plant-Data
AB  - Real-time quality monitoring in palm oil refining is often hindered by laboratory delays, which can be mitigated through the development of Artificial Neural Network (ANN) soft sensors. However, the effectiveness of these models is constrained by the characteristics of available process data. To address this, this study establishes a quantitative framework for assessing data set suitability by analyzing industrial data from three refineries using coefficient of variation (CV) statistics and self-organizing map (SOM) visualizations, with a feedforward neural network (FFN) used as a baseline model. Results show that process variables with CV exceeding 50% consistently yield better FFN performance than those below 50%. An integration strategy between different refineries was introduced to overcome the limitation of single-plant data and improve model performance to above 80% prediction accuracy for FFA content and Red Lovibond color. To test whether this improvement resulted from model capacity rather than data variability. The results further showed that data variability significantly affects model performance rather than model architecture. These findings establish CV as a practical criterion for data set suitability in ANN modeling for palm oil refining, offering an approach other chemical process industries facing similarly low-variation data may adapt.
SN  - 2470-1343
ER  -