Production Decision Data Assets

English | 简体中文

Applies to DecisioWorks v1.4.0

Production-Decision Data Assets: Value Begins with Computable Business Data

Manufacturers often already retain orders, products, routes, equipment, calendars, inventory, and execution records. These records become reusable production-decision data assets when they have stable business meaning, complete object relationships, and an explicit quality status that supports planning, scheduling, analysis, and integration.

Here, a data asset means a governed and reusable production-decision deliverable: standard templates, field mappings, object relationships, unit definitions, quality reports, and run evidence. The term describes engineering and operational value; accounting recognition and financial valuation are outside its scope.

Why data preparation already creates value

Starting condition	Data asset created	Work it supports
Data is split across ERP, MES, WMS, spreadsheets, and machine records	Sources, business objects, and field meanings are mapped	Clear ownership and less repeated interpretation
Fields exist, but product, route, resource, and material relationships are incomplete	Time, product, process-capacity, and material chains are validated	Earlier discovery of missing links, conflicts, and inconsistent definitions
Units, states, and defaults depend on personal knowledge	Units, enumerations, assumptions, and owners are recorded	Review, handover, and controlled updates
Files can be read, but solving readiness is unknown	ContractReady, ModelReady, and SolverReady become acceptance findings	A clear choice between repairing data, adding rules, or running a model
Every project rebuilds similar mappings	Templates, mapping rules, and test samples are retained	Reuse across scenarios, integrations, and AI Agent access

Even before production deployment, these deliverables show what current data can express, what is missing, who maintains it, and how far the scenario is from decision use.

From records to decision assets

1. Identify business objects

Identify orders, order lines, products, routes, operations, work centers, capacity, calendars, materials, inventory, and readiness relationships before mapping source columns. This prevents one field from carrying different meanings across systems.

2. Establish relationships and measurement conventions

Place delivery, shifts, capacity, and planning results on a consistent time axis. Connect products, routes, operations, and resources. Bind material demand to the relevant product, operation, and consumption point. Define units for quantity, time, ratio, and state.

3. Accept data in stages

• ContractReady: structure, fields, and basic semantics are identifiable;
• ModelReady: business objects, relationships, and structural constraints can construct a model;
• SolverReady: data, objectives, and runtime parameters can enter the real solving chain;
• DecisionReady: results, units, applied rules, and object relationships support business review and decision-making.

These names describe staged acceptance, consistent with the framework in Data readiness and validation. They are not a fixed set of states returned by every API. Relationship checks also need the separate RelationReady acceptance step. SolverReady means the selected entry point's solving prerequisites have been met; a real run and business review must still establish feasibility and suitability for execution. Order-level review requires a link or allocation between orders and results.

Each state is a risk finding that directs the next investment.

4. Validate assets through real runs

A real run tests whether the data can support planning and scheduling: demand closure, route continuity, capacity units, material relationships, and result traceability. Findings return to mappings, relationships, and rules, improving the asset over time.

What a transferable data asset contains

A minimum package usually includes:

• scenario scope, planning level, time unit, and data version;
• source-to-standard business-object mappings;
• table, field, enumeration, unit, and relationship definitions;
• data-quality, relationship-closure, and readiness reports;
• redacted samples, test methods, and known limitations;
• baseline configuration, run status, and object-level evidence;
• data owners, update methods, and change records.

Together, these materials support model validation, integration, training, troubleshooting, and scenario expansion.

How DecisioWorks supports the path

DataSets and data.db provide industry scenarios and standard-interface templates. data_layer maps objects and validates relationships. Scenario Lab exposes data, relationships, and readiness. GOCK and DecisioCore use real execution to test whether the data supports planning, scheduling, analysis, and feedback.

The data matures through an identifiable, model-ready, solver-ready, and review-ready path. The manufacturer can obtain a governed asset and a clear risk finding before deciding how far to expand models, scenarios, and system connections.

Continue with Standardized Data Interface, Data Readiness and Validation, and Typical Value Entry Scenarios.