Capital expenditure in water and wastewater infrastructure is, at its heart, a decision problem. How much to spend. Where to spend it. What technology to use. How to phase investment across time and geography. What tradeoffs to accept between cost, performance, and risk. And how to make all of these decisions under conditions of significant uncertainty, with regulatory deadlines, investor requirements, and community needs all pressing simultaneously. 

For Brazilian utilities and concessionaires operating under the 2033 universalisation targets, these decisions are being made at an unprecedented scale and pace. With 33 million Brazilians still lacking water access and 80 million without sewage treatment, the investment programme required to close these gaps is immense. Roland Berger estimates a BRL 550 billion investment requirement over the decade, with current investment rates requiring doubling to meet the 2033 timeline. Getting CAPEX allocation right, avoiding stranded investment, over-specified facilities, and technology choices that do not age well, is not just operationally important. It is financially critical. 

Generative design is emerging as one of the most powerful tools available to utilities for improving the quality of CAPEX decisions. Not because it is a financial tool, but because it brings engineering precision to the stage of the planning process where investment choices are actually made. 

Where CAPEX Waste Happens 

Before addressing how generative design helps, it is worth being precise about where CAPEX waste typically occurs in water infrastructure planning. 

The most common source of waste is not poor execution. It is poor specification at the early stages of project development. Facilities that are oversized relative to actual demand. Treatment configurations that are technically correct but not optimal for local conditions. Site layouts that create unnecessary civil complexity. Technology choices that are familiar rather than fit-for-purpose. These decisions are made at the conceptual and preliminary design stage, when engineering data is thin and time pressure is high. Once made, they are expensive to reverse. 

A second source of waste is insufficient optioneering. Most utilities, under time and resource constraints, evaluate a small number of design alternatives before committing to a preferred option. The alternative that gets chosen is not necessarily the best one available. It is often simply the one that could be evaluated within the available time and budget. 

A third source is misaligned technology selection. When the technology options available to a planning team are constrained by what they know well, what they have used before, or what suppliers have presented most persuasively, the resulting design may not reflect the range of genuinely viable options. Innovative technologies that could deliver better performance at lower cost, or with a smaller footprint, may simply not be evaluated because the evaluation process is too cumbersome. 

How Generative Design Changes the Equation 

Generative design addresses all three sources of CAPEX waste directly, by enabling planning teams to evaluate more options, faster, with greater engineering precision, than manual methods allow. 

On the optioneering front, platforms like the Transcend Design Generator can explore 18 alternative design scenarios in roughly 40 minutes, according to SABESP’s experience. Each scenario produces engineering-quality outputs: process flow diagrams, equipment lists, civil quantities, CAPEX and OPEX estimates, carbon footprint data, and land footprint analysis. This is not planning-level estimation. It is preliminary engineering at planning speed, with the level of analytical depth needed to make genuine investment decisions. 

On technology selection, generative design platforms that incorporate multiple technology options, including those integrated through partner programmes like Transcend Nexus, allow utilities to compare conventional and innovative treatment technologies on a genuine apples-to-apples basis. The comparison is made on engineering grounds, using accurate performance and cost data, not on the basis of familiarity or sales relationships. 

And on specification quality, the automated engineering logic embedded in a generative design platform applies consistent, rule-based decision-making across every design it generates. It does not skip steps under time pressure. It does not default to over-specification as a risk management strategy. It generates designs that are genuinely calibrated to the project’s requirements, not padded to create comfortable margins. 

A Real-World Example: BRK Ambiental 

The value of generative design for CAPEX optimisation is not theoretical. BRK Ambiental’s adoption of TDG produced documented, measurable results. The company reduced conceptual design costs by 80% and cut design timelines from two months to one week. Critically, this was not just a productivity improvement. It was a quality improvement: BRK’s engineering team found that TDG’s automated design logic, applied consistently across projects, produced more rigorous outputs than manually generated designs that varied based on individual engineer judgment and time availability. 

For a utility operating at BRK’s scale, with projects across multiple states and facility types, the compounding effect of more rigorous conceptual design across a large project portfolio is significant. Decisions that are better grounded in engineering reality at the conceptual stage are less likely to require costly revisions later. Facilities that are correctly specified at the planning stage are less likely to be oversized, under-performing, or misaligned with actual demand. 

CAPEX, OPEX, and the Whole-Life Cost Perspective 

One dimension of CAPEX optimisation that is frequently underweighted in traditional planning is the relationship between capital cost and whole-life operating cost. A technology or configuration that appears less expensive at the point of construction may carry higher energy costs, chemical costs, or maintenance requirements over a facility’s 30-year life. Conversely, a higher-specification technology may deliver lower operating costs that justify the additional capital outlay. 

Generative design platforms that model both CAPEX and OPEX simultaneously allow utilities to evaluate design options on a whole-life cost basis from the earliest stages of planning. This changes the investment decision from a question about what something costs to build, to a question about what it costs to build and operate over the facility’s full operational life, a much more financially meaningful frame for capital allocation decisions. 

For Brazilian concessionaires, who are accountable to investors over concession periods of 25 to 35 years, the whole-life cost perspective is not just analytically preferable. It is commercially essential. Investment decisions that optimise for short-term CAPEX at the expense of long-term OPEX erode the financial performance of the concession over time. 

Building the Analytical Capacity to Decide Well 

Ultimately, CAPEX optimisation is about the quality of investment decision-making. Generative design improves that quality by giving planning teams access to better data, more options, and faster analysis than manual methods can provide. 

For Brazilian utilities navigating the largest infrastructure build-out in the sector’s history, under intense time pressure and regulatory scrutiny, the ability to make investment decisions that are grounded in engineering reality, and to make them quickly, is one of the most important competitive and operational capabilities they can develop. 

The tools to develop that capability exist. The question is whether the sector’s leaders will prioritise building it before the investment decisions that define Brazil’s water infrastructure for the next generation have already been made. 

To learn how Transcend helps utilities generate engineering-quality design options for capital planning decisions, visit transcendinfra.com/capitalplanning.