Research synopsis

From BPMN models to smart contracts

How DRAK turns a business-analyst's process model into deployed, multi-chain smart contracts — no Solidity required.

The approach

A blockchain application is expressed by a business analyst using BPMN (control flow) and DMN (decision rules). DRAK transforms the model into a multi-modal Discrete-Event Hierarchical State Machine (DE-HSM) — time as discrete events, functionality as hierarchical state machines — and then into the methods of one or more smart contracts. Synchronization, cryptographic security and the transaction mechanism are generated automatically, so developers code only isolated task bodies.

Off-chain optimisation

Using SESE subgraphs of the DE-HSM model, DRAK finds patterns that are isolated from other concurrent activity and can run on a cheaper sidechain. Each pattern's mainchain cost is compared with its sidechain cost plus cross-chain overhead; cheaper patterns are deployed off-chain.

Nested ACID trade transactions

Long-term, collaborative activities spanning many methods are supported as nested trade (sub)transactions with Atomicity, Consistency, Isolation and Durability enforced by a generated mechanism — sub-transactions can be packaged as separate contracts on different chains for privacy.

Upgrade & repair (TABS+R)

When an unanticipated event prevents completion (e.g. a flood washes out a rail line), the failure is localised to the innermost sub-transaction; the modeller amends that BPMN fragment under pre/post-repair conditions, a new contract version is generated and deployed, the API is repointed, and execution resumes — preserving previously completed activities.

SCaaS payment rails

Payments are not modelled in BPMN; generic payment tasks are bound at runtime to pre-deployed payment services from a repository (on-chain native, cross-chain stablecoin USDC/USDT, off-chain card / wallet), with settlement-netting that aggregates many payments into a single transfer.

See the publications →