AI & ML
A Model-Driven Automation Pipeline for Building Operation
arXiv cs.SE0 views
arXiv:2611.21631v1 Announce Type: new Abstract: Building Control Systems (BCS) engineering in the Architecture, Engineering, Construction & Operations (AECO) domain remains a complex, manual, and error-prone process, with fractured tools and manual commissioning causing design and built reality to diverge into a performance gap. We argue that a fully integrated and automated, tool-supported workflow spanning design, programming, and operation is feasible and can improve efficiency while reducing manual effort and configuration errors. Building on model-driven engineering, we present an automated process pipeline for modeling, programming, and auto-commissioning BCS. It uses two domain-specific modeling languages with tooling support to synthesize and deploy executable artifacts from platform-independent models, replacing a fractured, multi-stakeholder handoff process with a defined four-step workflow and improving consistency across the engineering lifecycle. Our evaluation reports notable efficiency gains, improved adaptability, moderate-to-high user acceptance, and good performance with solid scalability.
Source:https://arxiv.org/abs/2611.21631PreviousReusing Battle-tested Code: A Systematic Literature Review of Automated Software TransplantationNextFragmented Markers, Mixed Results: A Systematic Review of AI Coding Assistants and Developer Productivity
Related posts
RoboKit: Accelerating Parallel Differentiable Optimization for Robot Learning
0arXiv cs.RO
Cleanly-DINO: Efficiently Learning Visual Task Spaces for Robotics
0arXiv cs.RO
SAFER: Navigating the Exploration-Exploitation Dilemma in World-Model-Based Planning for Autonomous Driving
0arXiv cs.RO
PlanDiv: Test-Time Scaling with Latent Plans for Vision-Language-Action Models
0arXiv cs.RO
MoViC: Real-Time Humanoid Whole-Body Control from Monocular Video Streams
0arXiv cs.RO
Practice Is Not Free: Measuring the Embodied Cost of Reset-Free Robot Practice
0arXiv cs.RO