Overview

In traditional manufacturing ecosystems, every device's capability description is fragmented: different manufacturers use different formats, different platforms adopt different interfaces, and different regions follow different standards. TCDP's mission is to eliminate these barriers, allowing every smart device to declare to the world in a unified, verifiable, and searchable way: I am here, I can do this, and here is how you can trust me.

Core Concepts

Capability Declaration

Each device publishes a structured capability declaration via TCDP, covering its manufacturing abilities (processable materials, precision, working area), operational status (online/offline/busy), and interface specifications (API endpoint, protocol version). This declaration is not a static document but continuously updated live data—automatically refreshed when device status changes, ensuring global information accuracy.

Semantic Schema

TCDP defines a hierarchical semantic schema that describes manufacturing capabilities from coarse-grained ("I can cut metal") to fine-grained ("I can cut steel up to 20mm thick with 0.05mm precision") in progressive levels. The schema supports extension—industry alliances can define their own sub-schemas while maintaining compatibility with the base schema.

Trust Credentials

Capability declarations must be accompanied by trust credentials: certificates issued by the Prometheus CA, reputation scores, and historical fulfillment records. These credentials ensure that declarations are not empty claims—any node can verify whether another device truly possesses its claimed capabilities and whether it has reliably completed past tasks.

Use Cases

• Global Device Discovery: Task initiators can search for devices worldwide that meet specific capability requirements without relying on any intermediary platform.
• Automatic Capability Matching: After decomposing tasks, the DTOP protocol automatically queries the TCDP registry to find the most suitable tool nodes.
• Cross-Vendor Collaboration: Devices of different brands and models can seamlessly participate in the same production network as long as they follow TCDP.
• Capability Evolution Tracking: Device upgrades, modifications, and degradations are all reflected in TCDP declarations, providing real-time global awareness.

Specification Details

TCDP declarations are expressed in JSON format and include the following mandatory fields:

• tool_id — Globally unique device identifier
• name — Device name
• type — Device type classification
• capabilities — Structured capability description (following semantic schema)
• location — Geographic location (latitude, longitude, and region)
• interfaces — API endpoint and protocol version
• trust — Trust credentials (certificate issuer, serial number, reputation score)
• status — Current operational status
• updated — Last update timestamp

For a complete TCDP declaration example, see the laser cutter example.

Overview

Traditional production scheduling relies on centralized systems: a factory's management software determines all process arrangements. DTOP breaks this paradigm. When a task enters the network, it is automatically decomposed into independently executable subtasks. Each subtask is broadcast globally, qualified tool nodes participate in bidding, and the task initiator dynamically selects the optimal combination based on capability, location, reputation, and price.

Task Decomposition

DTOP's decomposition engine breaks complex production requirements into atomic subtasks. Each subtask carries explicit capability requirements (from the TCDP semantic schema), time constraints, and quality standards. Decomposition follows the minimum dependency principle—dependencies between subtasks are explicitly declared but do not rely on a central scheduler to arrange execution order.

For example, "manufacture a metal gear" can be decomposed into:

• Subtask 1: Steel cutting (requirement: can cut 20mm steel)
• Subtask 2: Precision milling (requirement: 0.05mm precision, depends on Subtask 1)
• Subtask 3: Heat treatment (requirement: can reach 800°C, depends on Subtask 2)

Bidding Mechanism

After each subtask is broadcast globally, tool nodes with matching capabilities (verified via TCDP declarations) can submit bids. Bids include:

• Estimated completion time
• Execution price (based on the PCSP value system)
• Current load and available time slots
• Historical fulfillment records and reputation scores

The task initiator evaluates all bids holistically and selects the optimal plan. The bidding process is fully transparent—all participants can verify the rationality of the selection.

Dynamic Composition

DTOP supports dynamic composition—if a node executing a subtask fails midway (device malfunction, network disconnection), the system automatically triggers re-bidding to find a replacement node across the entire network, ensuring the task chain does not break due to a single point of failure. This is the core advantage of decentralized production: no single point of dependency means no single point of failure.

Trust Model

DTOP's trust model builds upon TCDP's trust credentials and further introduces:

• Fulfillment Deposits: Nodes lock a certain amount of value credentials when bidding; defaults result in forfeiture.
• Progressive Trust: New nodes start at a low trust level and gradually advance with increasing successful fulfillment counts.
• Network Consensus Verification: Physical work proofs after task completion are verified by multiple independent nodes, preventing false declarations.

Overview

In a decentralized production network, there is no company to issue paychecks and no bank to process transfers. PCSP provides an automated value settlement mechanism: every completed physical work automatically generates a contribution proof; every verified contribution proof automatically triggers a value settlement. The entire process requires no human intervention, no third-party trust, and is entirely protocol-driven.

Proof Mechanism

PCSP's contribution proofs differ fundamentally from blockchain's "Proof-of-Work" (PoW). PoW is artificially manufactured mathematical puzzles that consume computing power without producing real value. PCSP's contribution proofs are anchored in real physical work:

• Process Proof: Devices continuously generate cryptographically signed process data during task execution (sensor readings, operation logs, energy consumption records).
• Result Proof: After task completion, the physical output carries verifiable digital fingerprints (dimensional measurements, material analysis, quality inspections).
• Multi-party Verification: Contribution proofs are verified by at least three independent nodes, preventing forgery and tampering.

Settlement Flow

PCSP's settlement process is fully automated:

1. The task initiator pre-locks settlement deposits when submitting a task.
2. The tool node locks fulfillment deposits upon winning a bid.
3. After subtask completion, the node submits a contribution proof.
4. Verification nodes independently audit the contribution proof.
5. Upon verification success, deposits settle automatically: the initiator's payment releases to the executing node, and the executing node's fulfillment deposit is refunded.
6. Verification failure triggers a dispute arbitration process.

Value Exchange

PCSP defines a value unit system that enables fair comparison and exchange of different types of work:

• Base Value Unit: Anchored to standard processing duration and standard material consumption, defining "1 unit of processing value."
• Precision Premium: Work with higher precision requirements receives higher value multipliers.
• Time Premium: Urgency premiums for time-sensitive tasks are naturally determined by market supply and demand.
• Reputation Premium: High-reputation nodes' services enjoy market-recognized value additions.

Fairness Guarantees

PCSP ensures settlement fairness through multiple mechanisms:

• Transparent Pricing: All bidding and settlement records are publicly accessible and auditable by any participant.
• Multi-party Consensus: Verification results are jointly confirmed by multiple independent nodes—no single node can manipulate outcomes.
• Dispute Arbitration: Verification disagreements trigger randomly selected arbitration committees of neutral nodes.
• Anti-manipulation Protection: Value units and settlement rules are hard-coded in the protocol and cannot be unilaterally modified by any participant.

Live Demo

Below is a real running protocol backend instance. Click the button to watch how a task is decomposed, how tools bid, and how the settlement ledger updates. All data is computed live on the server.