designing-mes-systems

Design MES architectures, capability maps, and plant-system integration boundaries.

Updated Nov 22, 2023
One-click install
npx skills add https://github.com/parksanghoon-sys/TestCode --skill designing-mes-systems-parksanghoon-sys
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: designing-mes-systems
Source: https://github.com/parksanghoon-sys/TestCode/tree/main/src/Modbus/.agents/skills/designing-mes-systems
Command: npx skills add https://github.com/parksanghoon-sys/TestCode --skill designing-mes-systems-parksanghoon-sys

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Designing a Manufacturing Execution System requires clear decisions about what MES owns versus ERP, APS, WMS, QMS, LIMS, SCADA, and PLC systems, plus explicit modeling of orders, WIP, genealogy, and exception states. This Skill turns a manufacturing problem statement into a bounded, implementable, rollout-aware MES design instead of a vague platform diagram. ## Core Features & Use Cases - System Boundary Definition: Assigns ownership of business rules, master data, and state transitions across ERP, APS, WMS, QMS, LIMS, SCADA, PLC, and historian systems using the integration-boundaries reference. - Capability Mapping and Scoping: Builds a capability map covering order dispatch, route execution, WIP visibility, material genealogy, quality execution, and performance visibility, scoped by release. - Execution and Exception Modeling: Defines canonical objects, lifecycle states, and failure modes such as duplicate equipment events, network loss, and manual overrides using the execution-model checklist. - Use Case: A manufacturer wants to replace spreadsheet-based shop-floor tracking with MES. Use this Skill to produce a boundary matrix, first-release capability scope, state model for hold/rework/scrap flows, and a phased rollout plan with coexistence and fallback paths. ## Quick Start Use the designing-mes-systems skill to design an MES architecture for a single-site discrete manufacturing plant, including the system boundary with ERP and SCADA, the first-release capability map, and a phased rollout plan.

Frequently Asked Questions about designing-mes-systems

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I design an MES architecture for a manufacturing plant?▼

Start by framing the production mode (discrete, batch, process, or hybrid), then draw the MES system boundary against ERP, SCADA, and other systems before proposing modules. Build a capability map, model canonical objects and states, choose a runtime architecture, and plan a phased rollout.

What should MES own versus ERP, WMS, and SCADA?▼

MES typically owns dispatch, execution state, WIP truth, genealogy, and line-side material control. ERP owns orders, costing, and finance; WMS owns warehouse inventory; SCADA owns supervisory visualization and alarms. The integration-boundaries reference details ownership and interface direction for each system.

How do I scope the first release of an MES implementation?▼

Prioritize the capability that closes the most expensive manual loop or is required for traceability and compliance. The capability-map reference provides release-scoping prompts and warns against over-scoping traps like starting with scheduling optimization before dispatch discipline exists.

How should MES handle equipment connectivity failures and offline operation?▼

Designs should not assume always-online connectivity. Use store-and-forward queues at the site edge with operator-visible offline modes, make completion and consumption events idempotent, and define reconciliation rules for manual overrides performed outside MES.

What are common MES design anti-patterns to avoid?▼

Common anti-patterns include letting both MES and ERP decide operation completion, letting SCADA own traceability rules, sending raw high-frequency telemetry into MES instead of a historian, and decomposing into many microservices without proven deployment or autonomy value.