The Royal Institute of British Architects (RIBA) Plan of Work provides a structured framework for managing building projects from initial planning through construction and occupancy. Divided into eight stages, numbered 0 through 7, it defines responsibilities, deliverables, and approval requirements throughout the building lifecycle, says Pirros.

Stages 0 and 1 establish project objectives, feasibility, budgets, and responsibilities. Stage 2 develops the architectural concept, while Stage 3 coordinates architectural, structural, and building services designs. Planning applications are generally submitted after Stage 3.

Stage 4, Technical Design, is particularly demanding because architects must produce detailed construction drawings, specifications, schedules, and regulatory documentation. Stage 5 covers manufacturing and construction, followed by handover in Stage 6 and post-occupancy evaluation in Stage 7.

A recurring challenge during technical design is the unnecessary recreation of architectural details. Firms frequently redraw standard components, including window connections, roof junctions, and wall assemblies, because previously approved details are difficult to locate. Design decisions made during earlier stages may also become inconsistent with construction documentation.

Pirros addresses these inefficiencies through AI-powered detail management and quality control tools. Its platform searches existing architectural details, including Revit content, enabling teams to reuse established solutions instead of recreating them. Frequently used details can become standardized resources across projects.

Pirros Quality Management reviews drawings against organizational standards and identifies potential inconsistencies before documentation is issued.

The company’s Mira AI agent further automates repetitive Revit tasks, including replacing annotations, renumbering drawing details, and checking formatting consistency. Importantly, Mira presents proposed changes for user approval before modifying models.

The article emphasizes that effective project management requires consistent information across every stage. Feedback from construction and building operation should also inform future design standards.

By combining structured workflows with AI-assisted documentation, architecture firms can reduce repetitive work, improve quality control, and preserve valuable engineering knowledge across projects.

Real estate feasibility studies help developers determine whether a proposed project is financially viable and technically achievable. However, traditional assessments involving zoning regulations, building massing, parking requirements, and construction costs can take weeks and cost thousands of dollars. AI-powered software is accelerating these processes while improving early-stage decision-making.

This Snaptrude article compares seven feasibility study platforms: Snaptrude, TestFit, Deepblocks, ArchiWise AI, Digital Blue Foam, Higharc, and Northspyre. Each addresses different aspects of real estate development, from site selection and design optimization to financial planning.

Snaptrude combines AI-assisted zoning analysis, three-dimensional massing, and building information modeling (BIM). Its Universal Agent interprets setbacks, height restrictions, and floor area ratios to generate preliminary building envelopes. Developers can track floor areas, unit counts, and parking requirements as designs evolve. The resulting model can continue into architectural development.

TestFit specializes in generative site planning, parking layouts, and development yield optimization. Deepblocks analyzes zoning regulations, market conditions, and financial projections across hundreds of U.S. cities, helping developers identify promising properties.

ArchiWise AI supports early site screening by combining parcel information, zoning requirements, and financial feasibility assessments. Digital Blue Foam integrates environmental considerations, including carbon emissions, energy performance, wind, and solar exposure, into building massing studies.

Higharc focuses on residential development, connecting home design with construction estimates, code checks, and permit documentation. Northspyre addresses financial management through budgeting, cost forecasting, and project tracking.

Despite these capabilities, AI-generated feasibility assessments require professional verification. Zoning interpretations, development assumptions, and financial projections must remain transparent, editable, and traceable.

The article emphasizes that software selection should reflect project requirements rather than feature counts. Developers evaluating multiple properties may prioritize automated screening, while architects need tools that connect feasibility studies with detailed design.

Ultimately, AI can accelerate feasibility analysis, but professional judgment remains essential for evaluating development risks and ensuring projects are practical and financially sound.

Artificial intelligence is transforming urban infrastructure management by helping municipalities identify problems, prioritize maintenance, and make better use of limited resources. At a recent AI CityXchange workshop in Dublin, approximately 80 government technology leaders from Ireland and the United Kingdom explored practical applications of AI in city management, tells Bentley.

One example comes from Dún Laoghaire-Rathdown, near Dublin, where nine engineers oversee approximately 500 miles of roads, 850 miles of footpaths, and 60 miles of trails. Traditional inspections required months of manual assessment. AI-powered computer vision now analyzes images captured by vehicle-mounted cameras to identify road defects, damaged footpaths, and other infrastructure problems. This approach enables engineers to prioritize repairs and move toward predictive maintenance.

AI is also supporting environmental management. Engineering consultancy Mott MacDonald used stereo cameras and machine learning to catalog more than 13,000 ash trees for Glasgow City Council. The technology helps identify diseased trees, assess potential hazards, and estimate timber availability.

In Tipperary, Ireland, AI and geospatial technology are helping authorities address vacant commercial properties. The Vacant Business Reopen Grant platform combines property locations, planning information, and nearby amenities to identify redevelopment opportunities. Generative AI assists with reports and business proposals. More than 320 properties have been mapped.

A common element across these applications is geolocation, which connects infrastructure assets with relevant datasets. Emerging technologies such as Model Context Protocol (MCP) and Agent2Agent (A2A) could further simplify information sharing between AI systems and municipal applications.

Bentley Systems also highlights its Blyncsy platform, which uses AI to analyze anonymized dashboard-camera imagery for road monitoring.

However, increasing AI adoption raises questions about accountability, privacy, and governance. Municipalities must establish clear responsibilities as AI agents become more involved in operational decisions.

Ultimately, AI offers cities a practical way to improve infrastructure management, anticipate problems, and allocate resources more effectively.