Dr. Fajaruddin Bin Mustakim’s research focuses on intelligent transportation systems, vehicle-to-vehicle (V2V) communication, traffic behavior analysis, and advanced engineering simulation. His work integrates IR4.0 technologies and data-driven modeling to improve construction management, smart mobility, and sustainable infrastructure systems.

Fajaruddin Bin Mustakim, educational background includes the achievement of a degree in Bachelor of Civil Engineering from Universiti Teknologi Mara (UiTM) (1999), and subsequently receiving an MSc in Construction Management from Universiti Teknologi Malaysia (UTM) (2006). In 2014, he was awarded a PhD Engineering in the field of Scientific and Engineering Simulation from the Nagoya Institute of Technology (NITech), Japan. After completed PhD, continue his service at Universiti Tun Hussien Onn Malaysia (UTHM) as Senior Lecturer. My first post-doctoral position at Institute IR4.0 Universiti Kebangsaan Malaysia (UKM). Simultaneously has been appoint as consultant project entitle: TMR Asynchronous V2V with NLoS Vehicular Sensing (V2V) at MMU. This was then followed another position as academic collaborator at DatSINI Lab, Institute IR4.0, Universiti Kebangsaan Malaysia. Recently, he works at Malaysia Multimedia University (MMU) as Post-Doctoral Fellow. He has been active for many years in the fields of construction, transportation, and traffic behavior studies. In addition to this, he has authored at least 47 publications comprising of 21 indexed journal articles (ISI), 24 refereed conference articles, 3 edited books. He has supervised 2 PhD, 3 Master, 14 BSc and 2 Diploma. His strong belief is that the application of invention, inspiration and desire is necessary in anything that people want achievement in, and he believes that this is what ensures that academic work has a significant contribution to community and benefit to the mankind. Reference: Consequence of Accident Frequency and Driving Behaviour at Heterogeneous Traffic Flow Condition by Adopting Automated Driving Application, Machine Learning, and Neuron Network Simulation In this study, we developed the traffic behaviour model related to accident frequency (AF), and other traffic variables by using automated driving application (APA), binary logistic regression (BLR) and artificial neuron network (ANN) in urban area. Recently, neuron network fitting (NNF), Nonlinear Autoregressive Exogenous (NARX), and network pattern recognition (NNPR) were among the famous Machine Learning used in estimation process. The dataset was divided into three portions 70%, 15% and 15%, respectively, for learning validation and testing. The study, at first analyse the traffic behaviour at the selected unsignalized intersection (S) and make a comparison between those method NNF, NARX and NNPR to measure the model performance. Three techniques involved in the simulation process are Levenberg-Marquardt Algorithm (LMA), Bayesian Regularization Optimization (BRO) and Scaled Conjugate Gradient Algorithm (SCGA) to measure the MSE and tested the values. This study reveals, that NARX BRO and NNF BRO, performed the best result of mean square error of 0.15 and 0.08 respectively or accuracy performed at 85% and 92%. Moreover, the highest accident frequency often occurred typically during peak hour from (16:00-18:00). Meanwhile, from the BLR model, intersection providing the road safety infrastructure (RSI) encourage the right-turn motor vehicle (RMV) to accept longer gap. KeyWords: Machine Learning, Binary Logistic Regression (BLR), Neuron Network Fitting (NNF), Nonlinear Autoregressive Exogenous Model (NARX), And Network Pattern Recognition (NNPR).