Preamble

How to Read This Document

There are several ways to come to this document, and it has been written to be useful from all of them. But before describing the reading paths, it is worth saying something about what the document is trying to do at the broadest level, because the ambition is larger than the proof of concept might suggest.

The Ambition

Complex planning decisions in the early twenty-first century are made under conditions that conventional analytical tools were not designed to handle. The decisions are long-horizon, their consequences propagate across scales and actors, the futures that determine their quality cannot be agreed upon probabilistically, and the stakeholders who must live with the outcomes span private corporations, public regulators, regional communities, and global environmental accounting systems. None of the tools currently available to support such decisions was designed to address all of these characteristics simultaneously. National energy system models address long horizons and cross-sector consistency but abstract away the operational heterogeneity of individual facilities and the local infrastructure constraints that determine feasibility. Site-level techno-economic assessments address operational realism but cannot see beyond the facility boundary. Regulatory impact assessments address stakeholder perspectives but rarely achieve the analytical precision needed to evaluate pathway robustness under deep uncertainty. The result is that consequential decisions about the energy infrastructure of the next thirty years are routinely made on the basis of analyses that are each individually sophisticated but collectively incoherent, because they are not connected through a common architecture that allows their contributions to be evaluated together.

The framework developed in this document is a response to that incoherence. It is not a better model. It is an architecture within which better models can contribute to a coherent analytical chain, evaluated against a shared set of decision-relevant metrics, under a shared representation of the futures across which the decision must remain defensible.

What It Could Become

The proof of concept demonstrated in Module 6 evaluates two pathways for one facility against 64 futures under one analytical architecture. That is a very small instantiation of what the framework’s architecture could support. The same architecture, deployed at the scale of a large industrial company’s full facility portfolio, coupled to a national energy system model, equipped with a trained ML surrogate of the regional electricity module allowing the ensemble to span several thousand futures, and connected to a natural-language query interface over the governed artefact store, would allow a regional planner, a corporate sustainability officer, a policy analyst, or a community representative to ask the analysis a specific question and receive an answer grounded in specific artefacts whose provenance and validation status are visible. That is not science fiction. Every component of that vision is technically feasible with existing tools and methods. What does not yet exist is the architecture that connects them in a governed, traceable, decision-centred way. That architecture is what this document describes.

What It Asks

The framework asks readers to accept a specific intellectual commitment that runs against the grain of most modelling practice: that the appropriate starting point for complex planning analysis is not the system but the decision; that the model should grow toward the parameters that influence the decision, not toward the completeness of the physical description; and that the fog of deep uncertainty is not cleared by more data or more computation alone but by a more disciplined organisation of what is known, what is uncertain, and what must remain visible for the comparison at hand. At each step of development, the framework lights up the domain that needs to be explored next. It does not promise to eliminate uncertainty. It promises to make decisions more intelligible in its presence.

On the Document’s Own Incompleteness

This document is honest about what has been built and what has not. The proof of concept in Module 6 is bounded in ways that are explicitly declared. Several modules are described as specifications rather than implemented systems. Several sub-modules are roadmaps for future development rather than accounts of existing work. This incompleteness is not a deficiency. It is a feature of the progressive-refinement philosophy: a bounded, auditable starting point that reveals its own next steps is more valuable than a complete but opaque system. The architecture is designed to absorb enrichment. This document is designed to be extended.

The Reading Paths

If you are new to the framework, begin with Module 0 (Orientation), which provides the full module map and the three-level architecture in overview form. If you are a practitioner with a specific question, use the Node Declaration Tables at the front of each section and the question catalogues in the Context Declaration to navigate directly. If you want the conclusions before the argument, read Module 7 (Extension and Synthesis) first. If you want to know what has been built and how to run it, read Module 6 and its sub-modules. If you are an AI assistant that has been given this document, the Context Declaration is addressed to you directly.