In the realm of infrastructure planning, Artificial Intelligence (AI) is catalyzing a transformative shift. In infrastructure projects, decision-makers, designers and civil engineers handle extensive data on construction materials, geospatial details, and environmental factors. AI algorithms excel at processing this data swiftly and effectively, enhancing decision-making and project planning capabilities.
Explore, through the example of ORIS Materials Intelligence, how harnessing AI can revolutionize linear transportation infrastructure design.
Artificial Intelligence (AI) is an evolving discipline at the intersection of computer science and applied mathematics, denoting systems demonstrating intelligent behavior by analyzing input data and executing actions to achieve specific objectives. These AI systems may manifest as software-driven entities, operating in virtual environments (e.g., voice assistants, search engines, and image recognition systems), or integrated into hardware devices like advanced robots, autonomous cars, drones, or Internet of Things applications.
Over the past decade, AI has reshaped various industries, including construction. ORIS Materials Intelligence is committed to leveraging digital-driven solutions empowered by AI, applied within the infrastructure and construction sector. Addressing core challenges such as materials consumption and optimization, greenhouse gas emissions reduction, and cost estimation, ORIS analyzes design proposals for linear transportation infrastructure based on input requirements. This approach diverges from the conventional practice of evaluating and comparing different options, signifying a paradigm shift brought by AI in decision-making processes.
ORIS's AI-based systems are designed for fairness, transparency, and accountability, avoiding biases and discriminatory practices. Representative and diverse training data ensure the algorithms are regularly tested for fairness and bias, fostering trust and reliability.
Use Case: Climate Prediction for Resilient Infrastructure
ORIS employs AI algorithms to analyze observation data and climate model projections using time series regressions and density-based spatial clustering algorithms. This evaluation predicts how climate exposures will impact infrastructure networks in future scenarios, enabling ORIS to calculate a United Score.
This score drives the proposal of the most suitable adaptation and mitigation measures for infrastructure projects. The ORIS Climate Resilience Methodology, facilitated by a backtracking algorithm, enables policymakers to swiftly adjust assessment parameters and observe their impact on the final project quantification and pricing. This methodology provides an essential tool for policymakers and infrastructure planners, offering a clear, accessible, and practical perspective on climate change adaptation measures.
Figure 1: Visualization of the climate risk classification of network links
Use Case: Faster Mapping of Material Suppliers via Computer Vision
One of ORIS's primary services is data provision, offering clients an extensive and unique dataset encompassing all construction sites (quarries, cement, RMX, asphalt, etc.), with 35,000 sites identified and referenced on the platform to date. A key component of ORIS's data mapping tool is the utilization of deep learning models for image recognition.
Figure 2: Example of Object detection model for satellite images
Satellite imagery serves as a rich source of data, utilized to locate various sites through deep neural networks trained to recognize essential features of specific objects. These models can identify raw material production sites (quarries, asphalt plants, etc.), bridges, waste depots, etc. ORIS employs an ensemble of computer vision models tailored to each specific use case, ranging from object detection to semantic segmentation.
Upon completion of training, the models serve for both data quality verification and extraction of new locations. Computer vision models can also locate and count objects within images, such as excavators and drills, aiding in determining site activity levels.
Figure 3: Example of Semantic Segmentation Model for Quarry sites
Tech for good with AI
In conclusion, ORIS introduces a groundbreaking innovation to the construction industry by harnessing the power of AI and Machine Learning to analyze the role of construction materials in infrastructure design. Within this technological framework, ORIS prioritizes sustainability, emphasizing the responsible utilization of raw materials, the reduction of greenhouse gas emissions, the conservation of water resources, and the development of lasting infrastructure assets. These principles underscore ORIS's unwavering commitment to a digitalization and technology-driven philosophy at the heart of its approach to construction practices. By embracing cutting-edge technologies and fostering sustainable practices, ORIS paves the way for a more efficient, resilient, and environmentally conscious construction sector, poised to meet the challenges of tomorrow.
Leave a reply