CASE STUDIES
ATTIC Inc.
01
Predictive Modeling for Manufacturing & Service
In manufacturing processes, we identify the root causes of pain points and propose effective solutions to our clients. We also develop analytical models and optimize processes based on these models to enhance productivity, improve quality, and reduce costs across multiple domains. At ATTIC, we strive to help our clients gain a deeper understanding of their business through data-driven insights, enabling them to discover new opportunities and gain a competitive edge.
02
Image Classification, Recognition & Segmentation
A service that boosts operational efficiency by analyzing images faster and more accurately than humans.
Powered by deep learning–based computer vision algorithms, it performs tasks such as object detection, quantity analysis, defect inspection, and classification—handling repetitive or previously impossible work. From wastewater treatment systems to defect inspection systems, it delivers a new level of innovation across industries.
03
Natural Language Processing (NLP)
A service that provides language models capable of understanding human language—text or speech—quickly and accurately.
Built on pre-trained models, it delivers domain-specific capabilities such as topic classification, intent analysis, sentiment analysis, similarity analysis, natural language inference, and named entity recognition. Using large-scale language models and minimal training data, we create effective, specialized models for industries including finance, telecommunications, manufacturing, and services.
04
Environment, Healthcare & Safety
Attic, a social venture, has successfully executed projects such as data-driven equipment failure / prediction modeling and the establishment of pre-warning systems for potential hazards in petrochemical processes. Moreover, they are actively involved in research across environmental, safety, and healthcare sectors, including the optimization of operations for biogas plants and wastewater treatment facilities, developing optimal operation modeling to mitigate pollutants like nitrogen oxides (NOx), and analyzing genomics-based Omics data utilizing deep learning technology
-
1. Polymer product property prediction modeling
-
2. Olefin process optimization through product property prediction
-
3. Xylene production forecasting to boost output
-
4. Polymer quality prediction and commercialization for higher production
-
5. Semiconductor process data–based quality anomaly detection
-
6. Raw material classification via clustering analysis
-
7. Pressure drop prediction in residue hydrotreating processes
-
8. Real-time feed composition prediction from operating variables
-
9. Regenerator temperature prediction in catalyst decoking
-
10. Business performance forecasting from sales and marketing data
-
1. AI vision–based high-efficiency water quality management
-
2. Microbial image analysis for water treatment anomaly detection
-
3. Predictive maintenance and anomaly detection for biogas plants
-
4. Bacterial DNA analysis via sequencing methods
-
1. Developed an anomaly detection model for the coagulation process in water purification plants
-
2. Developed a waste plastic sorting and monitoring solution using machine vision technology
-
3. Process anomaly detection and predictive maintenance modeling for biogas plants
-
4. Wastewater quality prediction using process variables (COD and effluent SVI)
-
5. NOx reduction through root cause detection and optimal boiler operation
-
6. Advanced predictive maintenance modeling for rotating equipment
-
7. Root cause analysis of alarm events in compression systems
-
8. Medical data analysis for clinical decision support system development
-
9. Pre-warning model development for potential risks in residue hydrodesulfurization (RHDS) processes
네이버 아이디로 로그인 
카카오 아이디로 로그인 
구글 아이디로 로그인