Access Official Technical PDF

Share Summary:LinkedInEmail

Platform-based design for smart buildings provides a structured route from application requirements to a physical implementation. Rather than engineering each building function as a separate, self-contained system, the approach organizes reusable components into design platforms and maps requirements through successive levels of detail. For automation engineers, system integrators, and building owners, this creates a framework for examining alternatives before committing to hardware, software, and installation choices.

The need for that structure comes from the way smart buildings combine equipment, sensing, control, data analysis, and occupant needs. Ventilation, lighting, and other applications may depend on overlapping information, such as occupancy. Yet systems are often supplied and deployed independently, limiting component reuse and making integrated design more difficult.

Why smart building design needs a more systematic approach

A conventional one-function, one-box approach can be straightforward for delivering an individual application, but it may duplicate sensors, software, and installation work when multiple applications are added. An application-stack approach can allow functions to share infrastructure and data. However, simply enabling that sharing does not necessarily provide a systematic way to compare possible designs or determine which combination best meets project requirements.

Design decisions also involve competing considerations. A project may need to balance cost, occupant comfort, energy use, and operational needs while accommodating a fragmented supply chain and technologies at different stages of maturity. Building functions can interact, and a choice made for one application may affect another. These factors make ad hoc integration challenging, particularly when a building must be adapted over time.

How platform-based design for smart buildings works

Platform-based design separates what a system must do from how it will be implemented. The design process uses libraries of components, models, and composition rules. A component can describe its inputs and outputs, behavior, configuration parameters, and extra-functional properties such as cost or performance. Rules define which components can be connected and under what constraints.

The methodology combines a bottom-up and a top-down process. In the bottom-up phase, designers create libraries by abstracting available components and defining how they can be composed. In the top-down phase, requirements are mapped onto those libraries, with design space exploration used to compare feasible alternatives. This “meet-in-the-middle” structure aims to keep the search manageable while allowing the design to be refined toward implementation.

Three levels of design

  1. Function design: High-level requirements are translated into a prototype built from virtual devices and function templates. A requirement such as occupancy-driven HVAC control can be represented at this level without yet fixing every sensor or controller.
  2. Module design: Virtual devices are refined into basic functions such as sensing, control, actuation, and data analytics. Designers can compare implementations and account for additional details, including operating behavior, configuration, and cost.
  3. Implementation design: The selected modules are instantiated on the target building infrastructure. This stage can include deployment through a building operating system (BOS), code generation, and decisions such as sensor placement and configuration.

Each level uses an associated platform: the virtual device platform, module platform, and implementation platform. Architecture templates define ways of composing components at different levels, while design rules help prevent invalid combinations. The separation also allows a module’s function to be considered independently of the specific building operating system or device driver used to realize it.

Reuse and design space exploration

Shared infrastructure is valuable when different applications need similar information. For example, occupancy data may inform both demand-controlled ventilation and occupant-responsive lighting. A shared source of occupancy information can reduce duplicated sensing and software effort, provided that the selected design satisfies the requirements of both functions.

Design space exploration broadens the comparison beyond simply reusing an existing component. For occupancy detection, the design library might represent camera-based, WiFi-based, PIR-based, and CO2-based alternatives as distinct virtual sensors. The design process can assess these options against project constraints and relevant design rules. The aim is not to assume one technology is always preferable, but to make alternatives visible and evaluate them within the specific design problem.

Libraries can also capture differences in implementation effort. Data analytics modules, for instance, can describe both training and testing behavior. The amount of data collection or calibration associated with an approach is relevant to deployment planning, alongside its technical behavior and cost. If no existing library component meets the requirements, the methodology allows a new component or rule to be added for future design work.

Applying the method to an HVAC retrofit

Consider a commercial building project intended to make an existing HVAC system more energy efficient and responsive to occupancy. The design begins with requirements from the building owner, such as comfort conditions, a cost budget, an energy objective, and a payback horizon. Building information, including the building layout and occupancy schedules, can provide context for evaluating candidate designs.

At the function level, the team identifies the required HVAC behavior and the virtual devices needed to support it. At the module level, it considers sensing, analytics, control, and actuation options, including whether components can serve more than one function. At the implementation level, the selected design is mapped to the building’s actual control infrastructure, with deployment and sensor placement subject to applicable system constraints.

Validation matters throughout this process. Components that appear suitable individually can interact in ways that undermine the overall function. For example, a lighting control system that relies on a camera-based occupancy sensor could fail to detect people entering a dark room if the camera depends on lighting conditions. Simulation or formal verification can help identify such interactions before the design is deployed, prompting another round of design exploration when needed.

Practical implications for project teams

Platform-based design does not remove the need for engineering judgment or accurate building information. Its value lies in structuring the decisions: defining requirements, representing components and their constraints, comparing implementation options, and refining the design in stages. Project teams can use this structure to make reuse opportunities explicit and to identify integration issues earlier than a deployment-only approach might allow.

For building owners and technical decision-makers, the practical implication is a more organized basis for evaluating system choices over a building’s lifecycle. For integrators and engineers, it provides a way to connect high-level functions with sensing, control, software, and infrastructure without treating every application as an isolated product. As building requirements and technologies change, modular libraries can also be extended to support new alternatives. The result is a disciplined design process that links smart building goals to implementable systems while keeping cost, performance, and system interactions in view.

Access Official Technical PDF

About The Author