# Alternative Fillers in Asphalt Concrete Mixtures: Laboratory Investigation and Machine Learning Modeling towards Mechanical Performance Prediction

^{1}

^{2}

^{3}

^{*}

## Abstract

**:**

## 1. Introduction

_{2}emissions related to cement manufacturing. For this reason, a valuable alternative that is both environment-friendly and mechanically sound should be found [15]. A successful substitution requires the complete characterization of the recycled material or industrial waste to be reused. The most critical parameters that could affect the overall performance of the AM must be determined, namely: filler particle shape and size, specific gravity, surface area, mineralogical composition, and the presence of potentially harmful fine material [16,17]. In this framework, several waste materials have already been analyzed as potential alternative fillers, such as: copper slag powder [18], bauxite residue [19], empty palm fruit bunch ash [20], and coffee husk ash [21].

_{2}coupled with high percentages of CaO result in a higher water resistance and reduced moisture-induced damage [23]. The possibility of achieving similar or even slightly better performance by replacing traditional fillers with waste materials and the simultaneous reduction in waste disposal problems both encouraged the present study. The study focused on the investigation of AMs prepared with several materials as alternative fillers instead of ordinary Portland cement (OPC), which was used as a reference. The mechanical characterization of rice husk ash (RHA), brick dust (BD), marble dust (MD), stone dust (SD), fly ash (FA), limestone dust (LD), and silica fume (SF) was carried out, together with an in-depth comparative evaluation. Furthermore, Marshall, indirect tensile strength (ITS), moisture susceptibility and Cantabro abrasion loss (CL) tests were performed on AMs prepared with these fillers to evaluate their influence on the corresponding mixture’s behavior. Four different filler contents were considered, ranging from 4% to 8.5% by volume of the mixture, with a step-size of 1.5%.

## 2. Materials and Methods

#### 2.1. Aggregate, Bitumen and Fillers

#### 2.2. CatBoost Model

#### 2.3. Data Augmentation

## 3. Results and Discussion

#### 3.1. Laboratory Results

#### 3.2. CatBoost Modeling Results

^{2}(Table 7). For each of the five target variables, the mean absolute percentage error was always lower than 5%, with Pearson correlation coefficients higher than 0.95 and coefficients of determination higher than 0.88.

## 4. Conclusions

- Based on chemical and physical characterization, all the investigated alternative fillers could potentially be used in asphalt mixture design and replace OPC according to MoRTH standards;
- MoRTH requisites were fully satisfied also in terms of mechanical strength. Except for mixtures prepared with 4.0%, 5.5%, and 7.0% RHA and those prepared with 4.0% MD, all other mixtures showed a MS value above the acceptance threshold of 10 kN. LD and SF at 8.5% even overcame the acceptance requisite for modified bitumen mixtures, with MS values higher than 13 kN;
- All the investigated mixtures also satisfied MoRTH prescriptions in terms of moisture susceptibility, as they provided ITSR values consistently higher than 75%;
- In general terms, LD and SF were found to be the best alternative fillers in asphalt concrete mixtures among those investigated. They not only met MoRTH standards but even provided a better performance than OPC in terms of MS, ITS, ITSR, and CL;
- The data augmentation technique was worthwhile. Remarkable results in terms of all the evaluation metrics were obtained even on the basis of a small-size starting dataset;
- The CatBoost approach resulted in a successful predictive tool that can provide reliable mechanical performance predictions, thus avoiding expensive and time-consuming experimental procedures;
- The entire methodology was developed using Python 3.8.5. It is easily interpretable and implementable by other researchers. Any further application to datasets different from the one analyzed in this study requires new calibration and optimization of model hyperparameters.

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Data Availability Statement

## Acknowledgments

## Conflicts of Interest

## Abbreviations

AM | Asphalt Mixture |

AV | Air Voids |

BD | Brick Dust |

BF | Backhouse filler |

CL | Cantabro Loss |

CV | Categorical Variable |

FA | Fly Ash |

Fc | Filler content |

FM | Fineness Modulus |

ITS | Indirect Tensile Strength |

ITSR | Indirect Tensile Strength Ratio |

MBV | Methylene blue value |

MD | Marble Dust |

MF | Marshall Flow |

ML | Machine Learning |

MQ | Marshall Quotient |

MS | Marshall Stability |

LD | Limestone Dust |

OPC | Ordinary Portland Cement |

QD | Quarry Dust |

RHA | Rice Husk Ash |

SD | Stone Dust |

SF | Silica Fume |

VMA | Voids in the Mineral Aggregate |

VFB | Voids Filled with Bitumen |

## References

- Wang, L.; Gong, H.; Hou, Y.; Shu, X.; Huang, B. Advances in pavement materials, design, characterisation, and simulation. Road Mater. Pavement Des.
**2017**, 18, 1–11. [Google Scholar] [CrossRef] - Yilmaz, M.; Kök, B.V.; Kulolu, N. Effects of using asphaltite as filler on mechanical properties of hot mix asphalt. Constr. Build. Mater.
**2011**, 25, 4279–4286. [Google Scholar] [CrossRef] [Green Version] - Chen, M.; Lin, J.; Wu, S.; Liu, C. Utilization of recycled brick powder as alternative filler in asphalt mixture. Constr. Build. Mater.
**2011**, 25, 1532–1536. [Google Scholar] [CrossRef] - Veytskin, Y.; Bobko, C.; Castorena, C.; Kim, Y.R. Nanoindentation investigation of asphalt binder and mastic cohesion. Constr. Build. Mater.
**2015**, 100, 163–171. [Google Scholar] [CrossRef] - Dash, S.S.; Panda, M. Influence of mix parameters on design of cold bituminous mix. Constr. Build. Mater.
**2018**, 191, 376–385. [Google Scholar] [CrossRef] - Miró, R.; Martínez, A.H.; Pérez-Jiménez, F.E.; Botella, R.; Álvarez, A. Effect of filler nature and content on the bituminous mastic behaviour under cyclic loads. Constr. Build. Mater.
**2017**, 132, 33–42. [Google Scholar] [CrossRef] - Raposeiras, A.C.; Movilla-Quesada, D.; Bilbao-Novoa, R.; Cifuentes, C.; Ferrer-Norambuena, G.; Castro-Fresno, D. The use of copper slags as an aggregate replacement in asphalt mixes with RAP: Physical—chemical and mechanical behavioural analysis. Constr. Build. Mater.
**2018**, 190, 427–438. [Google Scholar] [CrossRef] - Xue, Y.; Hou, H.; Zhu, S.; Zha, J. Utilization of municipal solid waste incineration ash in stone mastic asphalt mixture: Pavement performance and environmental impact. Constr. Build. Mater.
**2009**, 23, 989–996. [Google Scholar] [CrossRef] - Ogundipe, O.M. Marshall Stability and Flow of Lime-modified Asphalt Concrete. Transp. Res. Procedia
**2016**, 14, 685–693. [Google Scholar] [CrossRef] [Green Version] - Lesueur, D.; Petit, J.; Ritter, H.J. The mechanisms of hydrated lime modification of asphalt mixtures: A state-of-the-art review. Road Mater. Pavement Des.
**2013**, 14, 1–16. [Google Scholar] [CrossRef] - Rashwan, N.K. Hot Mix Asphalt (HMA) Performance as Affected by Limestone Powder Filler Content. World Appl. Sci. J.
**2016**, 34, 237–244. [Google Scholar] [CrossRef] - Chrismer, J.L.; Durham, S.A. High Volume Fly Ash Concrete for Highway Pavements. In Proceedings of the Green Streets and Highways Conference, Reston, VA, USA, 14–17 November 2010. [Google Scholar] [CrossRef]
- Abdel-Wahed, T.A.; Rashwan, N.K. Application of Cement Dust and OPC as Mineral Filler in the binder Hot Mix Asphalt. In Proceedings of the 5th Annual International Conference on Asphalt, Pavement Engineering and Infrastructure, Liverpool, UK, 24–25 February 2016. [Google Scholar]
- Mazzoni, G.; Virgili, A.; Canestrari, F. Influence of different fillers and SBS modified bituminous blends on fatigue, self-healing and thixotropic performance of mastics. Road Mater. Pavement Des.
**2019**, 20, 656–670. [Google Scholar] [CrossRef] - Schneider, M.; Romer, M.; Tschudin, M.; Bolio, H. Sustainable cement production-present and future. Cem. Concr. Res.
**2011**, 41, 642–650. [Google Scholar] [CrossRef] - Wang, H.; Al-Qadi, I.L.; Faheem, A.F.; Bahia, H.U.; Yang, S.H.; Reinke, G.H. Effect of Mineral Filler Characteristics on Asphalt Mastic and Mixture Rutting Potential. Transp. Res. Rec.
**2011**, 2208, 33–39. [Google Scholar] [CrossRef] - Tiwari, N.; Satyam, N. Evaluation of Strength and Water Susceptibility Performance of Polypropylene Fiber-Reinforced and Silica Fume-Modified Hot Mix Asphalt. Adv. Civ. Eng. Mater.
**2021**, 10, 380–395. [Google Scholar] [CrossRef] - Modarres, A.; Alinia Bengar, P. Investigating the Indirect Tensile Stiffness, Toughness and Fatigue Life of Hot Mix Asphalt Containing Copper Slag Powder. Int. J. Pavement Eng.
**2019**, 20, 977–985. [Google Scholar] [CrossRef] - Choudhary, J.; Kumar, B.; Gupta, A. Performance Evaluation of Bauxite Residue Modified Asphalt Concrete Mixes. Eur. J. Environ. Civ. Eng.
**2019**, 26, 978–994. [Google Scholar] [CrossRef] - Lukjan, A.; Iyaruk, A.; Somboon, C. Evaluation on Mechanical Deterioration of the Asphalt Mixtures Containing Waste Materials when Exposed to Corrosion Solutions. Int. J. Eng. Technol.
**2022**, 12, 130–144. [Google Scholar] [CrossRef] - Tessema, A.T.; Wolelaw, N.M.; Alene, G.A. Experimental Evaluation of Coffee Husk Ash as a Filler in Hot Mix Asphalt Concrete Productions. Adv. Civ. Eng.
**2022**, 2022, 6726700. [Google Scholar] [CrossRef] - Awed, A.M.; Tarbay, E.W.; El-Badawy, S.M.; Azam, A.M. Performance characteristics of asphalt mixtures with industrial waste/by-product materials as mineral fillers under static and cyclic loading. Road Mater. Pavement Des.
**2022**, 23, 335–357. [Google Scholar] [CrossRef] - Hamedi, G.H.; Esmaeeli, M.R.; Najafi Moghaddam Gilani, V.; Hosseinian, S.M. The effect of aggregate-forming minerals on thermodynamic parameters using surface free energy concept and its relationship with the moisture susceptibility of asphalt mixtures. Adv. Civ. Eng.
**2021**, 2021, 8818681. [Google Scholar] [CrossRef] - Sholevar, N.; Golroo, A.; Esfahani, S.R. Machine learning techniques for pavement condition evaluation. Autom. Constr.
**2022**, 136, 104190. [Google Scholar] [CrossRef] - Hou, Y.; Li, Q.; Zhang, C.; Lu, G.; Ye, Z.; Chen, Y.; Wang, L.; Cao, D. The state-of-the-art review on applications of intrusive sensing, image processing techniques, and machine learning methods in pavement monitoring and analysis. Engineering
**2021**, 7, 845–856. [Google Scholar] [CrossRef] - Pasetto, M.; Baldo, N. Numerical visco-elastoplastic constitutive modelization of creep recovery tests on hot mix asphalt. J. Traffic Transp. Eng.
**2016**, 3, 390–397. [Google Scholar] [CrossRef] - De Oliveira Junior, M.; De Farias, M.M. A simple numerical methodology to simulate creep and recovery tests in HMA. Constr. Build. Mater.
**2020**, 262, 120793. [Google Scholar] [CrossRef] - Cao, P.; Leng, Z.; Shi, F.; Zhou, C.; Tan, Z.; Wang, Z. A novel visco-elastic damage model for asphalt concrete and its numerical implementation. Constr. Build. Mater.
**2020**, 264, 120261. [Google Scholar] [CrossRef] - Sun, B.; Hao, P.; Zhang, H.; Liu, J. Establishment and verification of a developed viscoelastic damage model and creep instability criterion for modified fine asphalt mortar. Mater. Struct.
**2022**, 55, 107. [Google Scholar] [CrossRef] - Pasetto, M.; Baldo, N. Computational analysis of the creep behaviour of bituminous mixtures. Constr. Build. Mater.
**2015**, 94, 784–790. [Google Scholar] [CrossRef] - Baldo, N.; Miani, M.; Rondinella, F.; Valentin, J.; Vackcová, P.; Manthos, E. Stiffness Data of High-Modulus Asphalt Concretes for Road Pavements: Predictive Modeling by Machine-Learning. Coatings
**2022**, 12, 54. [Google Scholar] [CrossRef] - Baldo, N.; Miani, M.; Rondinella, F.; Manthos, E.; Valentin, J. Road Pavement Asphalt Concretes for Thin Wearing Layers: A Machine Learning Approach towards Stiffness Modulus and Volumetric Properties Prediction. Period. Polytech. Civ. Eng.
**2022**, 66, 1087–1097. [Google Scholar] [CrossRef] - Upadhya, A.; Thakur, M.S.; Sihag, P.; Kumar, R.; Kumar, S.; Afeeza, A.; Afzal, A.; Saleel, C.A. Modelling and prediction of binder content using latest intelligent machine learning algorithms in carbon fiber reinforced asphalt concrete. Alex. Eng. J.
**2022**, 65, 131–149. [Google Scholar] [CrossRef] - Behbahani, H.; Hamedi, G.H.; Gilani, V.N.M. Predictive model of modified asphalt mixtures with nano hydrated lime to increase resistance to moisture and fatigue damages by the use of deicing agents. Constr. Build. Mater.
**2020**, 265, 120353. [Google Scholar] [CrossRef] - Gilani, V.N.M.; Hosseinian, S.M.; Behbahani, H.; Hamedi, G.H. Prediction and pareto-based multi-objective optimization of moisture and fatigue damages of asphalt mixtures modified with nano hydrated lime. Constr. Build. Mater.
**2020**, 261, 120509. [Google Scholar] [CrossRef] - Marcelino, P.; De Lurdes Antunes, M.; Fortunato, E.; Gomes, M.C. Machine learning approach for pavement performance prediction. Int. J. Pavement Eng.
**2021**, 22, 341–354. [Google Scholar] [CrossRef] - Zeiada, W.; Dabous, S.A.; Hamad, K.; Al-Ruzouq, R.; Khalil, M.A. Machine learning for pavement performance modelling in warm climate regions. Arab. J. Sci. Eng.
**2020**, 45, 4091–4109. [Google Scholar] [CrossRef] - Kandhal, P.S.; Lynn, C.Y.; Parker, F. Characterization Tests for Mineral Fillers Related to Performance of Asphalt Paving Mixtures. Transp. Res. Rec. J. Transp. Res. Board
**1998**, 1638, 101–110. [Google Scholar] [CrossRef] [Green Version] - Sousa, J.B.; Way, G.; Harvey, J.T.; Hines, M. Comparison of mix design concepts. Transp. Res. Rec.
**1995**, 1492, 151–160. [Google Scholar] - Pasandín, A.R.; Pérez, I. Overview of bituminous mixtures made with recycled concrete aggregates. Constr. Build. Mater.
**2015**, 74, 151–161. [Google Scholar] [CrossRef] [Green Version] - Al-Ammari, M.A.S.; Jakarni, F.M.; Muniandy, R.; Hassim, S. The effect of aggregate and compaction method on the physical properties of hot mix asphalt. IOP Conf. Ser. Mater. Sci. Eng.
**2019**, 512, 012003. [Google Scholar] [CrossRef] - Ozgan, E. Artificial neural network based modelling of the Marshall stability of asphalt concrete. Expert Syst. Appl.
**2011**, 38, 6025–6030. [Google Scholar] [CrossRef] - Zavrtanik, N.; Prosen, J.; Tušar, M.; Turk, G. The use of artificial neural networks for modeling air void content in aggregate mixture. Autom. Constr.
**2016**, 63, 155–161. [Google Scholar] [CrossRef] - Superpave Level 1 Mix Design, Superpave Series No. 2 (SP-2); Asphalt Institute: Lexington, KY, USA, 1995.
- D6927-15; Standard Test Method for Marshall Stability and Flow of Asphalt Mixtures. ASTM International: West Conshohocken, PA, USA, 2015.
- Zoorob, S.E.; Suparma, L.B. Laboratory Design and Investigation of the Properties of Continuously Graded Asphaltic Concrete Containing Recycled Plastics Aggregate Replacement (Plastiphalt). Cem. Concr. Compos.
**2000**, 22, 233–242. [Google Scholar] [CrossRef] - D6931-17; Standard Test Method for Indirect Tensile (IDT) Strength of Asphalt Mixtures. ASTM International: West Conshohocken, PA, USA, 2017.
- TxDOT: Tex-245-F Test Procedure for Cantabro Loss; Materials and Test Division Texas Department of Transportation: Austin, TX, USA, 2019.
- Prokhorenkova, L.; Gusev, G.; Vorobev, A.; Dorogush, A.V.; Gulin, A. Catboost: Unbiased boosting with categorical features. arXiv
**2017**, arXiv:1706.09516. [Google Scholar] [CrossRef] - Lou, Y.; Obukhov, M. Bdt: Gradient boosted decision tables for high accuracy and scoring efficiency. In Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD’17), Halifax, NS, Canada, 13–17 August 2017. [Google Scholar] [CrossRef]
- Bentéjac, C.; Csörgő, A.; Martínez-Muñoz, G. A comparative analysis of gradient boosting algorithms. Artif. Intell. Rev.
**2021**, 54, 1937–1967. [Google Scholar] [CrossRef] - Chen, T.; Guestrin, C. Xgboost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD’16), San Francisco, CA, USA, 13–17 August 2016. [Google Scholar]
- Ke, G.; Meng, Q.; Finley, T.; Wang, T.; Chen, W.; Ma, W.; Ye, Q.; Liu, T.Y. Lightgbm: A highly efficient gradient boosting decision tree. In Proceedings of the 31st International Conference on Neural Information Processing Systems (NIPS’17), Red Hook, NY, USA, 4–9 December 2017. [Google Scholar]
- Baldo, N.; Miani, M.; Rondinella, F.; Celauro, C. A Machine Learning Approach to Determine Airport Asphalt Concrete Layer Moduli Using Heavy Weight Deflectometer Data. Sustainability
**2021**, 13, 8831. [Google Scholar] [CrossRef] - James, G.; Witten, D.; Hastie, T.; Tibshirani, R. An Introduction to Statistical Learning with Applications in R; Springer: New York, NY, USA, 2013. [Google Scholar]
- Rahaman, M.S.; Liono, J.; Ren, Y.; Chan, J.; Kudo, S.; Rawling, T.; Salim, F.D. An ambient-physical system to infer concentration in open-plan workplace. IEEE Internet Things J.
**2020**, 7, 11576–11586. [Google Scholar] [CrossRef] - Oh, C.; Han, S.; Jeong, J. Time-Series Data Augmentation Based on Interpolation. Procedia Comput, Sci.
**2020**, 175, 64–71. [Google Scholar] [CrossRef] - Akima, H. A method of bivariate interpolation and smooth surface fitting based on local procedures. Commun. ACM
**1974**, 17, 18–20. [Google Scholar] [CrossRef] - Akima, H. A new method of interpolation and smooth curve fitting based on local procedures. J. ACM
**1970**, 17, 589–602. [Google Scholar] [CrossRef] - Ministry of Road Transport and Highways (MoRTH). Specifications for Road and Bridges Works, 5th Revision; Indian Roads Congress: New Delhi, India, 2013. [Google Scholar]
- Kutuk-Sert, T. Stability Analyses of Submicron-Boron Mineral Prepared by Mechanical Milling Process in Concrete Roads. Constr. Build. Mater.
**2016**, 121, 255–264. [Google Scholar] [CrossRef] - Nikolaides, A. Highway Engineering—Pavements, Materials and Control of Quality, 1st ed.; CRC Press Taylor & Francis Group: Boca Raton, FL, USA, 2015. [Google Scholar]
- Akbulut, H.; Gürer, C.; Çetin, S.; Elmaci, A. Investigation of using granite sludge as filler in bituminous hot mixtures. Constr. Build. Mater.
**2012**, 36, 430–436. [Google Scholar] [CrossRef] - Huang, B.; Shu, X.; Chen, X. Effects of mineral fillers on hot-mix asphalt laboratory-measured properties. Int. J. Pavement Eng.
**2007**, 8, 1–9. [Google Scholar] [CrossRef] - Arabani, M.; Tahami, S.A.; Taghipoor, M. Laboratory investigation of hot mix asphalt containing waste materials. Road Mater. Pavement Des.
**2017**, 18, 713–729. [Google Scholar] [CrossRef] - Kuity, A.; Jayaprakasan, S.; Das, A. Laboratory Investigation on Volume Proportioning Scheme of Mineral Fillers in Asphalt Mixture. Constr. Build. Mater.
**2014**, 68, 637–643. [Google Scholar] [CrossRef] - Pallant, J. SPSS Survival Manual: A Step by Step Guide to Data Analysis Using IBM SPSS; Open University Press: London, UK; McGraw-Hill: London, UK, 2016. [Google Scholar] [CrossRef]
- Saleh, R.; Fleyeh, H. Using supervised machine learning to predict the status of road signs. Transp. Res. Proc.
**2022**, 62, 221–228. [Google Scholar] [CrossRef]

**Figure 3.**Experimental data (black plus sign marker) and augmented data (red cross marker) for RHA mixtures.

**Figure 5.**Voids in the mineral aggregate vs. filler content for mixtures prepared with conventional and waste fillers.

**Figure 6.**Voids filled with bitumen vs. filler content for mixtures prepared with conventional and waste fillers.

**Figure 7.**Marshall stability vs. bitumen content for mixtures prepared with conventional and waste fillers.

**Figure 8.**Marshall quotient vs. filler content for mixtures prepared with conventional and waste fillers.

**Figure 9.**Indirect tensile strength vs. filler content for mixtures prepared with conventional and waste fillers.

**Figure 10.**Indirect tensile strength ratio vs. filler content for mixtures prepared with conventional and waste fillers.

**Figure 11.**Cantabro loss vs. filler content for mixtures prepared with conventional and waste fillers.

**Figure 15.**Test vectors and CatBoost predictions of MS (up-left), MQ (up-right), ITS (middle-left), ITSR (middle-right), CL (down).

**Figure 16.**CatBoost model regression plots for MS (

**up-left**), MQ (

**up-right**), ITS (

**middle-left**), ITSR (

**middle-right**), CL (

**down**).

Test Parameter | Method | Results | MoRTH Limits |
---|---|---|---|

Cleanliness (dust) (%) | IS 2386 Part I | 3 | 5 (Max) |

Bulk specific gravity (g/cm^{3}) | IS 2386 Part III | 2.68 | 2–3 |

Percent wear by Los Angeles abrasion (%) | IS 2386 Part IV | 10.6 | 35 (Max) |

Soundness loss by sodium sulphate solution (%) | IS 2386 Part V | 3.4 | 12 (Max) |

Soundness loss by magnesium sulphate solution (%) | IS 2386 Part V | 3.7 | 18 (Max) |

Flakiness and elongation index (%) | IS 2386 Part I | 35 (Max) | |

- -
- 20 mm
| 27.93 | ||

- -
- 10 mm
| 32.13 | ||

Impact strength (%) | IS 2386 Part IV | 27 (Max) | |

- -
- 20 mm
| 4.15 | ||

- -
- 10 mm
| 5.91 | ||

Water absorption (%) | IS 2386 Part III | 1.67 | 2 (Max) |

Test Parameter | Method | Results | MoRTH Limits |
---|---|---|---|

Absolute viscosity @60 °C (Poises) | IS 1206 (P–2) | 2855 | 2400–3600 |

Kinematic viscosity @135 °C (cSt) | IS 1206 (P–3) | 392 | 350 (Min) |

Flash point Cleveland open cup (°C) | IS 1448 (P–69) | 304 | 250 (Min) |

Penetration @25 °C, 100 g, 5 s, (1/10 mm) | IS 1203 | 49 | 45 (Min) |

Softening point (R&B) (°C) | IS 1205 | 48 | 47 (Min) |

Matter soluble in trichloroethylene (% by mass) | IS 1216 | 99.45 | 99 (Min) |

Viscosity ratio @60 °C | IS 1206 (P–2) | 1.3 | 4.0 (Max) |

Ductility @25 °C (cm after TFOT) | IS 1208 | 75 | 40 (Min) |

Specific gravity (g/cm^{3}) | IS 1202 | 0.987 | 0.97–1.02 |

Test Parameter | Mineral Filler Type | |||||||
---|---|---|---|---|---|---|---|---|

RHA | BD | MD | SD | FA | OPC | LD | SF | |

Specific gravity (g/cm^{3}) | 2.02 | 2.56 | 2.69 | 2.69 | 2.32 | 3.04 | 2.65 | 2.20 |

MBV (g/kg) | 4.72 | 6.25 | 4.45 | 3.67 | 3.86 | 3.00 | 3.75 | 3.85 |

German filler (g) | 65 | 40 | 70 | 85 | 75 | 85 | 97 | 94 |

FM | 3.21 | 5.17 | 2.12 | 5.38 | 3.77 | 4.96 | 3.03 | 1.96 |

Surface area (m^{2}/g) | 2.31 | 2.69 | 4.37 | 2.70 | 2.19 | 1.75 | 2.70 | 16.45 |

pH | 10.86 | 8.67 | 8.50 | 12.57 | 7.30 | 12.90 | 10.22 | 6.98 |

SiO_{2} (%) | 89.67 | 39.55 | 0.60 | 82.37 | 48.24 | 21.43 | 0.48 | 93.5 |

CaO (%) | 1.88 | 12.88 | 55.60 | 2.79 | 13.40 | 66.58 | 96.57 | 0.89 |

Al_{2}O_{3} (%) | 1.62 | 15.71 | 0.40 | 8.23 | 24.15 | 3.01 | 0.41 | 0.08 |

MgO (%) | 0.97 | 3.29 | 0.10 | 1.47 | 1.46 | 1.39 | 0.46 | 0.82 |

Fe_{2}O_{3} (%) | 1.06 | 14.05 | 0.20 | 5.27 | 6.48 | 4.68 | 0.32 | 0.50 |

Particle shape | Honeycombed | Subangular particles | Subangular particles | Angular particles | Rounded | Granular/ subangular particles | Granular particles | Spherically shaped |

Feature | Grid | Selected Value |
---|---|---|

Number of iterations | 100, 250, 500, 1000, 2000 | 1000 |

Max depth | 1, 2, 3, 4, 5 | 3 |

Learning rate | 0.5, 0.1, 0.05, 0.01, 0.005 | 0.1 |

k-fold Cross-validation | – | 4 |

Overfitting detector | – | 20 |

Loss function | – | MultiRMSE |

Metric | Definition | Analytical Expression |
---|---|---|

MAE | Mean absolute error | $\frac{1}{N}{\displaystyle {\displaystyle \sum}_{i=1}^{N}}\left|{y}_{{T}_{i}}-{y}_{{P}_{i}}\right|$ |

MAPE | Mean absolute percentage error | $\frac{1}{N}{\displaystyle {\displaystyle \sum}_{i=1}^{N}}\left|\frac{{y}_{{T}_{i}}-{y}_{{P}_{i}}}{{y}_{{T}_{i}}}\right|\times 100$ |

MSE | Mean squared error | $\frac{1}{N}{\displaystyle {\displaystyle \sum}_{i=1}^{N}}{\left({y}_{{T}_{i}}-{y}_{{P}_{i}}\right)}^{2}$ |

RMSE | Root mean squared error | $\sqrt{\frac{1}{N}{\displaystyle {\displaystyle \sum}_{i=1}^{N}}{\left({y}_{{T}_{i}}-{y}_{{P}_{i}}\right)}^{2}}$ |

R | Pearson correlation coefficient | $\frac{1}{N-1}{\displaystyle {\displaystyle \sum}_{i=1}^{N}}\left(\frac{{y}_{{T}_{i}}-{\mu}_{{y}_{{T}_{i}}}}{{\sigma}_{{y}_{{T}_{i}}}}\right)\left(\frac{{y}_{{P}_{i}}-{\mu}_{{y}_{{P}_{i}}}}{{\sigma}_{{y}_{{P}_{i}}}}\right)$ |

R^{2} | Coefficient of determination | $1-\frac{{{\displaystyle \sum}}_{i=1}^{N}{\left({y}_{{T}_{i}}-{y}_{{P}_{i}}\right)}^{2}}{{{\displaystyle \sum}}_{i=1}^{N}{\left({y}_{{T}_{i}}-{\mu}_{{y}_{{T}_{i}}}\right)}^{2}}$ |

Variable | Description | U.M. | Count | Mean | Std. Dev. |
---|---|---|---|---|---|

CV | Categorical variable–filler type | – | 56 | – | – |

Fc | Filler content | % | 56 | 6.25 | 1.51 |

AV | Air voids | % | 56 | 4.42 | 0.40 |

MS | Marshall stability | kN | 56 | 11.88 | 1.42 |

MQ | Marshall quotient | kN/mm | 56 | 3.93 | 0.74 |

ITS | Indirect tensile strength | kPa | 56 | 901.61 | 107.82 |

ITSR | Indirect tensile strength ratio | % | 56 | 85.35 | 3.76 |

CL | Cantabro Loss | % | 56 | 15.73 | 4.05 |

Metric | MS | MQ | ITS | ITSR | CL |
---|---|---|---|---|---|

MAE | 0.2595 | 0.1099 | 29.5495 | 0.6441 | 0.7317 |

MAPE | 2.3328 | 3.0829 | 3.4862 | 0.7669 | 4.3340 |

MSE | 0.1101 | 0.0195 | 1432.5907 | 0.6495 | 0.7670 |

RMSE | 0.3319 | 0.1396 | 37.8496 | 0.8059 | 0.8758 |

R | 0.9778 | 0.9826 | 0.9543 | 0.9744 | 0.9727 |

R^{2} | 0.9478 | 0.9626 | 0.8885 | 0.9446 | 0.9437 |

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |

© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Tiwari, N.; Rondinella, F.; Satyam, N.; Baldo, N.
Alternative Fillers in Asphalt Concrete Mixtures: Laboratory Investigation and Machine Learning Modeling towards Mechanical Performance Prediction. *Materials* **2023**, *16*, 807.
https://doi.org/10.3390/ma16020807

**AMA Style**

Tiwari N, Rondinella F, Satyam N, Baldo N.
Alternative Fillers in Asphalt Concrete Mixtures: Laboratory Investigation and Machine Learning Modeling towards Mechanical Performance Prediction. *Materials*. 2023; 16(2):807.
https://doi.org/10.3390/ma16020807

**Chicago/Turabian Style**

Tiwari, Nitin, Fabio Rondinella, Neelima Satyam, and Nicola Baldo.
2023. "Alternative Fillers in Asphalt Concrete Mixtures: Laboratory Investigation and Machine Learning Modeling towards Mechanical Performance Prediction" *Materials* 16, no. 2: 807.
https://doi.org/10.3390/ma16020807