A Design Technology Manager at one of the nation’s largest engineering and technical services firms explains why a mixed-capture strategy — terrestrial scanner, SLAM, drone, and phone — beats any single technology approach.
Tetra Tech is one of the largest engineering and technical services firms in the United States, a publicly traded company with operations spanning water, environment, energy, and infrastructure that employs roughly 28,000 people worldwide. Inside that sprawling enterprise, individual operating groups retain considerable identity and technical autonomy. Tetra Tech Rooney — the Colorado-based midstream oil and gas pipeline engineering group that Tetra Tech acquired in 2012 — is one such group, and it is where Chris Fries, Design Technology Manager, has spent the last several years quietly building one of the more disciplined reality capture and digital delivery programs in the pipeline sector.
Fries’s role spans technology selection, reality capture operations, BIM coordination, and emerging technology implementation for a team of roughly 100 people. The conversation he describes is one that most engineering firms are having right now: what do you actually do with all the data you’re collecting, and how do you make it useful to clients who weren’t hired to manage digital infrastructure?
Scan-to-BIM as the Baseline
The core workflow for Tetra Tech Rooney’s existing facilities work begins with reality capture — primarily using FARO terrestrial scanners, including a FARO Orbis premium — and moves through a stack of Autodesk products: Revit for BIM modeling, Navisworks for coordination and review, AutoCAD for 2D deliverables, and Autodesk ReCap for point cloud processing. The group holds two Autodesk Professional Network IDs, reflecting the depth of the Autodesk integration.
For most of the industry’s history, the deliverable coming out of that pipeline was the point cloud itself, maybe paired with a model. Tetra Tech Rooney has been systematically shifting away from point clouds as the primary client-facing artifact, leaning instead on segmented meshes — and for reasons that are more practical than philosophical.
“The biggest thing is file size,” Fries said. “Point clouds are great and robust, but they’re so big, they’re very difficult to share.” Getting a point cloud into Navisworks at usable performance levels remains a hardware challenge, and for many clients — particularly the midstream operators Tetra Tech Rooney serves — the data format itself creates comprehension barriers. A point cloud is semi-transparent, its density variable; a wall or a pipe rack doesn’t read as a wall or a pipe rack to someone unfamiliar with the representation.
Segmented meshes address both problems. At roughly one-tenth the file size of a comparable point cloud, they’re portable, shareable, and renderable on hardware that wouldn’t handle the raw scan data. They sacrifice some accuracy — Fries is direct about that — but for the spatial awareness use cases that dominate most client interactions, the tradeoff is sound. When bolt-level or flange-level accuracy is genuinely required, the workflow falls back to the point cloud and direct modeling. For everything else, the mesh gets the job done with less friction.
The client-facing value proposition has become increasingly clear. Rather than delivering a folder of static photographs and a PDF survey, Tetra Tech Rooney is giving clients an immersive Navisworks model — a combined point cloud, mesh, and photo environment — that allows site personnel, project managers, and contractors to walk the facility virtually. “You don’t have to go back to the site again,” Fries said. “You’ve captured more than you would have with a traditional survey and pictures. You’re creating a true virtual environment.” The safety and cost implications in a midstream oil and gas context, where remote or hazardous site visits are the norm, are significant.
Mixed Capture: The Toolbox Approach
The principle Fries comes back to most consistently is that no single capture technology is the right tool for every part of a job. Tetra Tech Rooney has developed what amounts to a tiered capture strategy, deploying different instruments for different accuracy requirements within the same project.
For a large facility — thirty acres of pipeline infrastructure, pump stations, containment dikes, and tank farms — a full terrestrial scan of every element would be time-consuming and expensive beyond what most accuracy requirements justify. Fries’s approach is to match the instrument to the specification. Where the engineering tie-in needs sub-inch accuracy — a manifold, a flange connection, a critical dimensional reference — the FARO terrestrial scanner goes in. Where the rest of the site needs to be captured for spatial context but doesn’t need millimeter-level precision, the Orbis SLAM scanner covers the ground faster and at lower cost. For current topography and aerial imagery across the whole site, a drone provides a coverage efficiency that ground-based instruments can’t match.
“We’re using three or four different technologies on one site,” Fries said. “Don’t be stuck with just one technology.” The return on that flexibility is captured data volume that a single-instrument approach would never achieve at comparable cost, combined with the accuracy where it actually matters.
The same logic extends to follow-up visits. If a site has changed and a specific area needs to be updated in the model, a 3D camera can be deployed for a targeted re-capture and incorporated back into the existing dataset, without requiring a full remobilization. The result is a living model that can be updated incrementally as the facility evolves.
Tetra Tech Rooney’s geomatics group — a separate team within the broader organization — extends the capability further with a heavy-payload LiDAR drone, commercial drone platforms for remote sensing, and now a bathymetric sensor for underwater data collection. Fries’s group draws on that resource for aerial coverage while focusing its own instrumentation on the ground-level and interior capture that defines most of the midstream plant work.
VR and the 30% Review
The virtual reality initiative Fries describes began roughly 18 months before the interview and addresses a specific pain point in engineering review workflows. Standard Navisworks desktop reviews — navigating a 3D coordination model on a monitor — require users to maintain and manipulate a mental model of spatial relationships that a flat screen doesn’t intuitively convey. Is there enough clearance between the new pipe run and the existing structure? Is the handle at the right height? Can personnel egress safely?
VR provides what desktop review cannot: true one-to-one spatial awareness. Walking a model at full scale, with depth perception intact, resolves spatial questions in a fraction of the time. Fries cited a roughly 30% reduction in review time compared to a conventional Navisworks session, alongside better review quality — the VR environment surfaces clashes and ergonomic issues that monitors tend to obscure.
A current pilot project at a Tetra Tech group in Canada illustrates the safety dimension. Bridge inspection traditionally involves personnel physically suspended over the structure. The Canadian group is using VR to conduct inspections from the model environment, keeping personnel off the bridge entirely. It’s a narrow application but a representative one: the value of immersive review scales with the danger and logistical complexity of the physical alternative.
AI: Grounded Expectations, Real Initiatives
Fries’s assessment of AI’s near-term impact is measured — he’s candid that broad productivity transformation is probably overstated in the short term — but he described a set of concrete, funded initiatives within Tetra Tech that point toward where the technology is genuinely taking hold.
Tetra Tech’s internal Delta Group functions as a centralized AI development unit for the broader organization, and it has built tools that are already in production. One is an internal chatbot for proposal writing and technical documentation — a focused, bounded application that accelerates a specific high-volume workflow. Two others came out of Tech 1000, Tetra Tech’s internal innovation competition, in which teams compete to develop client-facing solutions: an application for automated crack detection on building facades, and a runway crack analysis system. Both use machine vision on image data to identify structural defects at a speed and consistency human inspection can’t match.
The facade scanning application is particularly close to Tetra Tech Rooney’s daily work — the combination of reality capture and AI-assisted feature recognition for infrastructure inspection is a direct extension of the scan-to-BIM pipeline into a predictive maintenance use case. Fries’s group doesn’t run these AI tools directly, but the organizational awareness of what’s possible is shaping the direction of the technology roadmap.
Digital Twins: Early but Accelerating
The digital twin conversation in Tetra Tech Rooney’s client base is at an early but meaningfully active stage. Fries describes the current state of most of the group’s deliverables as “crude digital twins” — georeferenced point clouds, Revit models, Navisworks coordination files. The data is there; it’s not yet connected to live operational sensors or integrated into asset management platforms. The twin is static.
The more advanced work is happening elsewhere in the Tetra Tech organization. The RPS Group, Tetra Tech’s European engineering practice, is further along — developing live digital twins for government and defense clients, including work that Fries characterizes as highly classified. Other operating groups are working with Aveva for connected asset management and evaluating Esri products for GIS integration. Autodesk Tandem is on the roadmap for Tetra Tech Rooney’s own digital twin ambitions.
The client-side barrier is real. Most midstream oil and gas operators — Fries is frank about this — are not on the cutting edge of data management. They build and operate pipelines; platform-level asset intelligence is not their core competency, and they haven’t historically invested in the infrastructure to support live twins. What’s changing is the recognition that years of reality capture investment represent a data asset that isn’t being managed. “They’re spending all this money on capturing assets,” Fries noted, and the conversation about how to store, manage, and sustain that data over time is beginning in earnest. Exxon’s announced initiative to scan all its facilities and build digital twins is the kind of marquee signal that tends to move the rest of the industry.
The trend Fries projects over the next three years is a shift in what gets delivered to clients — away from PDF record drawings toward model-based delivery, where the Navisworks coordination model or the BIM file becomes the primary record document rather than an ancillary reference. Contractors who need dimensional information can work directly from the model. The PDF doesn’t disappear, but it stops being the deliverable of record.
Hardware and Software Driving Adoption
The overarching observation Fries offered on the state of the market is straightforward but worth stating clearly: reality capture adoption in the energy sector is accelerating primarily because the technology has become dramatically easier and cheaper to deploy, not because clients have become more technically sophisticated. Five or six years ago, selling a reality capture scope required educating a client on what they were buying and why it was worth the investment. Today, scanning existing facilities before starting engineering work has become close to standard practice for Tetra Tech Rooney’s clients. The value proposition — accurate as-built documentation, virtual site access, reduced field visits — has become self-evident.
The improvement has been hardware and software in parallel: scanners that capture faster and register more reliably, processing pipelines that turn raw data into usable models with less manual intervention, and cloud platforms that make the outputs accessible to people who aren’t specialists. AI will add another layer of efficiency as the tools mature. But the adoption curve Fries describes is already well underway, driven by instruments and software that have quietly gotten good enough to remove the friction that used to make the case hard to make.