Using multiple linear regression to predict engine oil life

J 2025

Using multiple linear regression to predict engine oil life

NGUYEN, Van Thai; Jan FURCH a Jan KOLÁČEK

Základní údaje

Originální název

Using multiple linear regression to predict engine oil life

Autoři

NGUYEN, Van Thai; Jan FURCH a Jan KOLÁČEK

Vydání

Scientific Reports, NATURE PUBLISHING GROUP, 2025, 2045-2322

Další údaje

Jazyk

angličtina

Typ výsledku

Článek v odborném periodiku

Obor

10102 Applied mathematics

Stát vydavatele

Velká Británie a Severní Irsko

Utajení

není předmětem státního či obchodního tajemství

Odkazy

URL

Impakt faktor

Impact factor: 3.900 v roce 2024

Označené pro přenos do RIV

Ano

Kód RIV

RIV/00216224:14310/25:00142181

Organizační jednotka

Přírodovědecká fakulta

Klíčová slova anglicky

engine oil life; multiple regression; Bayesian model averaging

Štítky

14311010, rivok

Příznaky

Mezinárodní význam, Recenzováno

Změněno: 11. 11. 2025 08:49, Mgr. Marie Novosadová Šípková, DiS.

Anotace

V originále

This paper deals with the use of multiple linear regression to predict the viscosity of engine oil at 100 °C based on the analysis of selected parameters obtained by Fourier transform infrared spectroscopy (FTIR). The spectral range (4000–650 cm⁻¹), resolution (4 cm⁻¹), and key pre-processing steps such as baseline correction, normalization, and noise filtering applied prior to modeling. A standardized laboratory method was used to analyze 221 samples of used motor oils. The prediction model was built based on the values of Total Base Number (TBN), fuel content, oxidation, sulphation and Anti-wear Particles (APP). Given the large number of potential predictors, stepwise regression was first used to select relevant variables, followed by Bayesian Model Averaging (BMA) to optimize model selection. Based on these methods, a regression relationship was developed for the prediction of viscosity at 100 °C. The calibration model was subsequently validated, and its accuracy was determined using the Root Mean Squared Error (RMSE) metric, it was 0.287. Finally, the obtained model was used to predict the lifetime of engine oil in diesel engines operating under severe conditions.

Návaznosti

MUNI/A/1569/2024, interní kód MU

Název: Specifický výzkum v odborné, aplikované a učitelské matematice 2025

Investor: Masarykova univerzita, Specifický výzkum v odborné, aplikované a učitelské matematice 2025

Citovat

NGUYEN, Van Thai; Jan FURCH a Jan KOLÁČEK. Using multiple linear regression to predict engine oil life. Scientific Reports. NATURE PUBLISHING GROUP, 2025, roč. 15, č. 33585, s. 33585-33598. ISSN 2045-2322. Dostupné z: https://doi.org/10.1038/s41598-025-18745-w.

@article{2522737,
   author = {Nguyen, Van Thai and Furch, Jan and Koláček, Jan},
   article_number = {33585},
   doi = {https://doi.org/10.1038/s41598-025-18745-w},
   keywords = {engine oil life; multiple regression; Bayesian model averaging},
   language = {eng},
   issn = {2045-2322},
   journal = {Scientific Reports},
   title = {Using multiple linear regression to predict engine oil life},
   url = {https://rdcu.be/eI2Ho},
   volume = {15},
   year = {2025}
}

TY  - JOUR
ID  - 2522737
AU  - Nguyen, Van Thai - Furch, Jan - Koláček, Jan
PY  - 2025
TI  - Using multiple linear regression to predict engine oil life
JF  - Scientific Reports
VL  - 15
IS  - 33585
SP  - 33585
EP  - 33585
PB  - NATURE PUBLISHING GROUP
SN  - 20452322
KW  - engine oil life
KW  - multiple regression
KW  - Bayesian model averaging
UR  - https://rdcu.be/eI2Ho
N2  - This paper deals with the use of multiple linear regression to predict the viscosity of engine oil at 100 °C based on the analysis of selected parameters obtained by Fourier transform infrared spectroscopy (FTIR). The spectral range (4000–650 cm⁻¹), resolution (4 cm⁻¹), and key pre-processing steps such as baseline correction, normalization, and noise filtering applied prior to modeling. A standardized laboratory method was used to analyze 221 samples of used motor oils. The prediction model was built based on the values of Total Base Number (TBN), fuel content, oxidation, sulphation and Anti-wear Particles (APP). Given the large number of potential predictors, stepwise regression was first used to select relevant variables, followed by Bayesian Model Averaging (BMA) to optimize model selection. Based on these methods, a regression relationship was developed for the prediction of viscosity at 100 °C. The calibration model was subsequently validated, and its accuracy was determined using the Root Mean Squared Error (RMSE) metric, it was 0.287. Finally, the obtained model was used to predict the lifetime of engine oil in diesel engines operating under severe conditions.
ER  -

NGUYEN, Van Thai; Jan FURCH a Jan KOLÁČEK. Using multiple linear regression to predict engine oil life. \textit{Scientific Reports}. NATURE PUBLISHING GROUP, 2025, roč.~15, č.~33585, s.~33585-33598. ISSN~2045-2322. Dostupné z: https://doi.org/10.1038/s41598-025-18745-w.

Přehled o publikaci