Discipline 03 / Reference

Digital Engineering

Digital engineering is the practice of maintaining an authoritative digital model of a system and using that model, rather than a document, as the basis for designing, deciding, and operating. The term comes from systems engineering, where the shift was from drawings to models. Applied to an organization, it means the process map, the automations built on it, and the telemetry watching it are the same artifact rather than three that drift apart. This page covers what a model-based approach replaces, how a process is actually captured, and which junctions stay under human control.

Field
Digital engineering, model-based systems engineering
Draws on
Systems engineering, process mining, knowledge representation, software architecture, human factors
Formalized
U.S. Department of Defense Digital Engineering Strategy, 2018
Central claim
The source of truth should be a maintained model, not a document that describes one
Primary methods
Process capture, BPMN 2.0 modeling, agentic execution with audit trail, causal telemetry
Applied here to
H-Synth platform, workflow automation, operational instrumentation
01

Documents drift. Models are maintained.

The distinction sounds administrative and is not. A document is a snapshot that begins decaying the moment it is signed, and nothing in its structure reports its own staleness. A model is connected to the thing it describes, so divergence between the two are more readily detectable, and often even automatic. Every difference below follows from that one.

PropertyDocument-basedModel-based
Source of truthA file, usually several files, usually disagreeing.One model that every downstream artifact is generated from.
How it gets updatedManually, by whoever remembers, during a review cycle.Continuously, from observed execution and from explicit change.
Detecting stalenessSomeone notices during an incident, or an audit finds it.Telemetry compares executed behavior against the model and flags divergence.
Automation built on itInherits whatever was wrong at the moment of writing, then scales it.Regenerated when the model changes, so drift is corrected rather than compounded.
Cost of a changeFind every affected document. Miss one.Change the model. Downstream artifacts follow.
Characteristic failureConfident execution of a version of the organization that no longer exists.Model neglect. The approach fails when nobody is accountable for maintenance.
02

Anatomy of a process map.

Process maps are written in BPMN 2.0, an open notation standardized by the Object Management Group in 2011. Using a standard rather than a house diagram style has two consequences worth stating plainly. The map is readable by other tools and by people who have never met us, and the same map can be executed rather than only consulted. The fragment below shows the four element types that carry most of the meaning.

Receive requestUNDER THRESHOLDApprove automaticallyOVER THRESHOLDRoute to reviewerCIRCLE / EVENTRECTANGLE / TASKDIAMOND / EXCLUSIVE GATEWAY
Figure 1. A minimal BPMN 2.0 fragment. The gateway is exclusive, meaning exactly one branch is taken. The threshold on that gateway is the interesting part of most real maps: it is where policy is actually encoded, it is frequently undocumented, and different people in the same organization often believe it sits at different values.
03

How a process gets captured.

Capture begins from what already exists rather than from a blank page, which is the difference between 4-5 business days, and the twelve to eighteen months a conventional mapping exercise consumes. Roughly a week per major workflow, at full task-level detail.

01IngestDocs, logs, recordings02ReconstructDraft the real flow03ValidateWith the people doing it04FormalizeBPMN 2.0 and graph05InstrumentSignals and audit trail
Figure 2. Stage three is the one that cannot be compressed. A map validated only against documentation reproduces the documentation, including everything already wrong with it.
  1. 01

    Ingest what already exists

    Standard operating procedures, policy manuals, training material, system logs, OpenAPI specifications, screen recordings, interview audio, even photographs of whiteboards. No greenfield requirement, and no request that anyone write documentation first.

  2. 02

    Reconstruct the flow as executed

    Machine learning over the ingested material produces a draft flow, with system logs carrying more weight than prose because logs record what happened rather than what was intended. Contradictions between sources are kept rather than resolved, since they mark where the real disagreement sits.

  3. 03

    Validate with the people doing the work

    Semi-structured interviews test the draft against practice and recover the tacit rules no source recorded. This is where undocumented thresholds, standing exceptions, and the informal escalation path get written down for the first time.

  4. 04

    Formalize into a standard notation

    The validated flow becomes BPMN 2.0 diagrams and an organizational knowledge graph linking processes, roles, systems, and decisions. Outputs also render as reports, checklists, and policy documents, in whatever format your team already uses.

  5. 05

    Instrument the running system

    Causal telemetry compares executed behavior against the model, so divergence surfaces as a signal rather than as an incident. Every automated execution writes a deterministic audit trail recording what ran, on what input, and under whose authority.

04

Where a person stays in the loop.

Human in the loop is a phrase that has been diluted to the point of meaning very little. Below is the specific version: five junction types where the system is constrained from acting alone, what it does instead, and what gets written to the record. The last column matters as much as the others, because an authority that leaves no trace cannot be audited and is therefore not really an authority.

JunctionWhat the system doesWhat the person doesWhat is recorded
ClassificationProposes a category and reports its confidence in that proposal.Accepts, overrides, or escalates.The proposal, the confidence, the decision, and who made it.
Threshold breachHalts and surfaces the case rather than applying the rule.Decides whether to proceed, and on what basis.The halt, the stated reasoning, the outcome.
Novel caseDeclines to act and flags the input as outside what it was built on.Handles it manually. The case is then added to the model.The decline, the manual resolution, the model update.
Irreversible actionNever executes autonomously, regardless of confidence.Authorizes explicitly, as a distinct act.The authorization and the authorizer.
Policy conflictSurfaces both applicable rules and names the conflict between them.Chooses, and the choice becomes precedent.The conflict and the precedent set by resolving it.
05

What this produces, and what it does not.

Automation projects tend to fail on the second column rather than the first. Reading it before starting is cheaper than discovering it in month nine.

Produces
  • Structured, auditable process maps in BPMN 2.0, plus reports, checklists, and policy documents generated from the same model.
  • An organizational knowledge graph connecting processes to the roles, systems, and decisions they depend on.
  • A deterministic causal audit trail for every automated execution, showing what ran, on what input, and why.
  • A role-based access matrix defining who can view, decide, approve, and escalate at each junction.
  • Telemetry that reports divergence between the model and actual execution while it is still small.
Does not produce
  • A system that maintains itself. A model with no owner decays the same way a document does, only faster, because more depends on it.
  • Automated judgment. What gets automated is execution and retrieval. The consequential calls stay at the junctions above.
  • Value from a process the organization has not actually agreed on. Capture will expose the disagreement clearly and cannot settle it.
  • A replacement for the people holding context. Removing them removes the thing that made the model correct in the first place.
  • Vendor lock-in. Outputs are delivered in your required formats, and the model is portable by construction. That is a design constraint, not a concession.
06

Where the field came from.

Digital engineering is the newest of the three disciplines and the most institutionally driven. Systems engineering had been moving from documents toward models since the 1990s, and the U.S. Department of Defense made the shift policy in 2018, which pulled a large supplier base along with it. The intellectual debts run older: to process mining for recovering behavior from logs, and to safety engineering for the insight that accidents are properties of systems rather than failures of components.

  1. Object Management Group, Business Process Model and Notation, version 2.0 (2011). The specification the maps are written against. Open, tool-portable, and executable, which is why we use it rather than a house notation.
  2. Wil van der Aalst, Process Mining (2011). How to recover the process that actually ran from the traces it left in system logs. The methodological basis for reconstructing a flow without asking anyone first.
  3. Nancy Leveson, Engineering a Safer World (2011). The argument that accidents emerge from interactions between working components rather than from broken ones, and the control-theoretic model that follows from it.
  4. U.S. Department of Defense, Digital Engineering Strategy (2018). The policy document that made the authoritative model the expected practice across defense acquisition, and the reason the term has the currency it does.
  5. INCOSE, Systems Engineering Vision 2035. The professional body's account of where model-based practice is heading, and a reasonable statement of what the approach is expected to deliver.
  6. Lisanne Bainbridge, Ironies of Automation, Automatica (1983). Listed on the Cognitive Science page as well, and belonging here just as much. Automating the routine work makes the remaining work harder, which is a design problem rather than a training problem.