Page 1 of 9

Transactions on Engineering and Computing Sciences - Vol. 13, No. 02

Publication Date: April 25, 2025

DOI:10.14738/tecs.1302.18245.

Kim, B. S., Choi, J., & Le, S. Y. (2025). A Cognitive Analysis and Life Prediction Through AI Algorithm of Control Arm Using

Manufacturing and Vehicle Driving Data. Transactions on Engineering and Computing Sciences, 13(02). 10-18.

Services for Science and Education – United Kingdom

A Cognitive Analysis and Life Prediction Through AI Algorithm of

Control Arm Using Manufacturing and Vehicle Driving Data

Byeong Sam Kim

Departement of Automotive Engineering,

Hoseo University, Asan city, Korea

Jinuk Choi

R&D Center, Induswell co. Ltd, CheonAncity, Korea

Sang Yeoul Le

R&D Center, R&D Center, Hwangsung co. Ltd

ABSTRACT

This study aims to enhance vehicle safety by predicting the life perdition of control

arms, critical suspension components. Traditional inspection methods have

limitations in accurately predicting failures, leading to unexpected accidents

occurring both before and after the vehicle's expected lifespan. The increasing

complexity of control arm manufacturing, coupled with the growing volume of

vehicle driving data and heightened competition, necessitates a more sophisticated

approach to quality and safety. This system implements autonomy and intelligence

of the production system by utilizing intelligent production system, big data, and

artificial intelligence technologies, and supports optimal decision-making in real

time. Data collection: There collected various sensor data from the production site,

system data, MES system data, etc. In data refinement, data analysis and algorithm

extraction of the Control Arm are performed, and the collected data is refined and

preprocessed to be processed into a form suitable for analysis. Database

construction: We build a relational database or NoSQL database to systematically

manage data. This study represents a crucial step towards a more proactive and

data-driven approach to vehicle safety and manufacturing. By integrating AI and big

data technologies, the automotive industry can move towards a future

characterized by minimized accidents and optimized production processes. Finally,

we derived the results of predicting and optimizing the remaining useful life

prediction of the remaining product.

Keywords: Control arm, cognitive analysis detection technique, AI data analysis,

intelligent production system, life perdition, smart factory.

INTRODUCTION

The control arm, one of the automobile parts, is one of the core components of the suspension

system and acts as a direct connection point between the front wheel assembly and the vehicle

frame. It is a component used in the suspension of the control arm, and it connects the body and

the wheels to absorb the shock of the road surface. The control arm may look simple in

appearance, but it is one of the core components that plays an important role in the overall

stability and driving performance of the vehicle. The control arm, which can be seen in almost

Page 2 of 9

11

Kim, B. S., Choi, J., & Le, S. Y. (2025). A Cognitive Analysis and Life Prediction Through AI Algorithm of Control Arm Using Manufacturing and Vehicle

Driving Data. Transactions on Engineering and Computing Sciences, 13(02). 10-18.

URL: http://dx.doi.org/10.14738/tecs.1302.18245

all road-driving suspension systems, is located on the front axle of each of the two front wheels.

In the manufacturing process, the suspension control arm is made of cast iron, and the same

material was used in this study. The cast iron control arm is a component that provides

strength, rigidity, and damage resistance.

Fig 1: Installing the control arm system

The present invention relates to the control arm that constitutes the suspension of an

automobile, and in particular, to a manufacturing method of a control arm manufactured by

extrusion using a material such as aluminum, and to an extruded material and a control arm

manufactured by this manufacturing method. In the manufacturing process of the present

control arm, the control arm is manufactured by an extrusion process of extruding an aluminum

billet through an extrusion die equipped with an extrusion hole along the horizontal cross- sectional shape of a control arm to form an extruded material extending in a direction matching

the vertical height direction of a control arm to be manufactured, a cutting process of cutting

the extruded material in a direction perpendicular to the vertical height direction of the control

arm to be manufactured to form a plurality of primary intermediate materials, a processing

process of forming a joining portion for joining parts to the primary intermediate materials to

form a secondary intermediate material, and an assembly process of separately manufacturing

a spring seat cup on which a seat of a coil spring to be coupled to the control arm is to be placed,

and assembling the spring seat cup and the seat to the secondary intermediate material.

CONTROL ARM SYSTEM

Control Arm Manufacturing Proces

By installing it on the control arm, which is a suspension component, we attempted to predict

the remaining lifespan due to vehicle component defects and fundamentally prevent vehicle

accidents based on this. In the past, only simple inspections and tests were conducted, but now

we need a guarantee for accelerated life accidents that occur before and after the vehicle's

durability life. To this end, we individually installed complex sensors on important vehicle parts

to prevent accidents during driving in advance through non-contact sensors that cannot

physically interact with people, such as autonomous driving.

Fig 2: Control arm adjustment device and measurement position of each sensor

Page 3 of 9

12

Transactions on Engineering and Computing Sciences (TECS) Vol 13, Issue 02, April - 2025

Services for Science and Education – United Kingdom

Processing of Manufacturing Generated Data

In this study, we attempted to complete the component level by conducting real-time

measurement tests on three types of data (sensor/road/vehicle) through T, a suspension parts

manufacturer, and control arm parts equipped with sensors, and obtaining measurement data.

The acquired measurement data was filtered using the PRUL algorithm technology and

completed vehicle level development through field tests at the actual road test site of the Korea

Automobile Research Institute, and composite data for five life prediction sensors were

constructed. At this time, this study was conducted using the PRUL (Prognostics of Remaining

Useful Life) sensor and system on-device. The generated data was processed. As a comparative

study of this study, the prospects after the completion of the development of the technology are

quite encouraging, such as improving vehicle safety, securing safety solutions, and extending

the life of related parts by measuring the status of vehicle driving safety parts that have a

serious impact on driver safety and vehicle operation in real time during normal driving of the

vehicle, diagnosing failures related to the durability of the parts, and transmitting and

controlling preemptive prediction information. In particular, in the past, it was limited to simple

predictions based on inspection processes and accelerated life tests, or collecting and notifying

information after failures occurred, and in the case of further improvements, it was limited to

transmitting information on maintenance consumables such as tires/wheels/oil/brakes. In this

way, after the completion of the development of the technology, the initiative in information on

vehicle maintenance management was acquired, and based on this, there were similar contents

such as identifying responsibility and efficiently managing optimal maintenance of driving

parts other than consumables.

Autonomous Driving Safety Technology System for Automobiles

A comparative study examining the future technology system aspect of autonomous vehicles

shows that risks have been continuously raised, including the fatal accident caused by Tesla

malfunction in the US, 11 minor accidents involving Google's self-driving car in 6 years, and a

traffic accident on the first day of fully autonomous driving in the US. In order to ensure safety,

a driving platform based on multi-safety technology that can minimize the risk of accidents

even if a problem occurs in the vehicle is essential. For Level 0-2 vehicles, a fail-safe system that

detects and notifies a failure and relies on manual driver action is required, and for Level 3 and

higher, a fail-operational system based on multi-safety design that overcomes the failure itself

before driver intervention is required when a failure is detected. In particular, the functional

safety standard for automotive electrical components (ISO26262) recommends multi-mode

design, the International Forum for Harmonization of Automotive Standards (UNECE/WP. 29)

is reviewing strict safety requirement standards for autonomous vehicles, and the US National

Highway Traffic Safety Administration (NHTSA) announced architecture recommendations for

functional safety of major ADAS systems, and in particular, the steering system recommended

redundancy of power, sensors, and actuators. Accordingly, GM introduced multiple motors,

multiple controllers, and multiple power supplies for the steering and braking actuators in the

Cruise AV. Furthermore, the Safety Of The Intended Functionality (SOTIF) standard (ISO/PAS

21448), unlike functional safety (ISO26262), considers cases where the intended design itself

is insufficient to secure safety and is inappropriate rather than malfunctions, failures, and

defects.