Duration
June 2023 – March 2024
Case Study 02 · Developer Tools
SIMIT Rapid Tester — Engineering Automation Platform
Transforming manual validation into an automated, low-code engineering workflow.
- Role
- Lead Designer — End-to-End Design · Patent Co-author
- Domain
- Industrial Automation · Developer Tools
- Duration
- June 2023 – March 2024
- Tools
- Figma, Adobe XD, Maze
Overview
The strategic frame
I designed the tool that let automation engineers stop validating plant logic by hand — and trust an automated result instead, in an environment where a wrong pass reaches a real plant.
Outcome
75% shorter test cycles
Adoption
20+ customers since launch
Patent
2024E15630 IN
Problem
The problem and its stakes
Automation engineering in process industries is complex and changes constantly. Guaranteeing correctness means repeatedly testing control logic, interlocks and sequences — and that work was manual, done under time pressure, often before real hardware was available.
What was at stake
Four costs compounded: high effort per cycle, real risk of human error, limited test coverage, and mounting frustration with re-testing after every engineering change. Errors that escaped early testing surfaced later in the plant lifecycle, where they cost far more to fix.
Why it was hard
Trust had to be earned, not assumed
Engineers validated logic by hand because the stakes were physical. Asking them to delegate that to a tool meant the tool had to make its reasoning legible — what ran, what passed, what failed, and why.
Native was required, and that closed doors
Zero-latency coupling with simulation and DCS ruled out a web application. I designed a Windows app, but deliberately built the visual language to web conventions so a future web transition would not mean starting over.
Scale was extreme
A single coupling can expose millions of signals. Any interaction that worked at forty signals and broke at ten million was not a real solution.
Research
Evidence under constraints
Engineering workshops
A series of workshops with automation engineers working on simulation engineering, plus discussions with SMEs, produced the requirement set directly.
Integration constraints
The tool had to access systems under test — SIMATIC PCS 7, SIMATIC PCS neo, TIA Portal — over OPC UA, trigger user inputs and errors, and validate correct engineering behaviour.
Pattern precedent
The interaction model was first validated in an earlier SIMIT training-simulation tool, where stakeholder response proved the approach before it was adapted here.
What worked
- Low-code drag-and-drop modelling was adopted beyond the specialist group
- Type/instance concept scaled one template across many test instances
- Analytical reporting improved traceability across the plant lifecycle
Personas
User landscape
Automation Engineers
Core Developers- Context
- Engineering workstations, complex automation projects
- Needs
- Rapid test creation, automated validation, result analysis
- Goals
- Validate automation logic efficiently, reduce manual dev time
- Challenges
- Complex test scenarios, time pressure, safety
- Success metrics
- Development speed, validation accuracy, delivery time
Test Engineers
Validation Specialists- Context
- QA labs, systematic testing, validation protocols
- Needs
- Comprehensive coverage, detailed analytics, compliance docs
- Goals
- Ensure quality, maintain standards, generate reports
- Challenges
- Test complexity management, documentation requirements
- Success metrics
- Test coverage, defect detection rate, compliance
Project Managers
Delivery Focus- Context
- Project coordination, timelines, resource allocation
- Needs
- Progress visibility, resource planning, delivery predictability
- Goals
- Accelerate timelines, optimise resources, ensure quality
- Challenges
- Bottlenecks, allocation, predictability
- Success metrics
- Delivery speed, resource efficiency, quality
Engineering Leads
Strategic Oversight- Context
- Technical leadership, planning, team coordination
- Needs
- Productivity metrics, adoption, strategic insights
- Goals
- Improve efficiency, drive innovation, maintain quality
- Challenges
- Adoption, team productivity, strategic alignment
- Success metrics
- Efficiency improvement, innovation adoption, impact
Design judgement
Decisions under constraint
01
Engineers Decide the Blast Radius of a Failure
Problem
When a verification fails mid-run, the right response depends entirely on context — sometimes the run should continue, sometimes everything downstream is invalid. Choosing on the engineer's behalf would be wrong either way.
Solution
Every Verify action carries an explicit 'if verification fails' choice — Continue, Break group, Break instance, or Break test set — paired with an Important flag per action so engineers mark which assertions actually matter and filter results by that judgement.
Result
The engineer's expertise stays in control of the automation instead of being overridden by it.
02
The Workflow Teaches Itself Through Empty States
Problem
The tool has a mandatory sequence — connect couplings, fetch signals, design test types, create instances, build test sets, execute — and a wrong start produced confusing dead ends.
Solution
Every empty state names its prerequisite rather than just reporting emptiness, guiding users to create couplings before viewing signals, and to add a test type before designing test actions.
Result
The sequence becomes learnable without training material, answering 'what now?' at the moment the question arises.
03
Safe Bulk Editing at Scale
Problem
Renaming a signal across a large project is essential but risky — a careless global replace silently corrupts a validation suite across millions of signals.
Solution
Find & Replace shows a result count first, then a confirmation dialog with a scoped tree where engineers select exactly which folders and test types to change and see a live count before committing, followed by a success confirmation linked to a replacement log.
Result
Preview, scope, confirm, audit — a destructive operation made recoverable in practice.
04
Results That Explain Themselves
Problem
A pass/fail verdict alone doesn't build trust; engineers need to see what the tool actually did.
Solution
Results expand step by step with each action's description, execution comment, importance flag and outcome, and connection attempts are logged as first-class results so an environment failure is never mistaken for a logic failure.
Result
Engineers can audit the automation rather than take it on faith — the precondition for handing over work they previously did by hand.
05
Low-Code Validation Architecture
Problem
Complex programming made test-sequence creation inaccessible to broader engineering teams.
Solution
A visual workflow builder and drag-and-drop interface supported rapid scenario creation.
Result
Automation testing became accessible to broader engineering teams.
06
Automated Validation Framework
Problem
Validation needed to run in an automated testing environment while keeping results understandable.
Solution
Integrated validation logic with intelligent error detection and diagnostic recommendations.
Result
The framework provided simple result interpretation within the automated testing environment.
07
Analytics-Driven Insights
Problem
Engineering teams needed both an overview and detail when reviewing test information.
Solution
Interactive dashboards supported movement between overview and detail.
Result
The analytics surfaced optimisation and efficiency opportunities.
08
Virtual Testing Environment
Problem
Testing depended on live systems despite complex industrial scenarios.
Solution
A virtual environment supported automation testing without live system dependencies.
Result
Testing could use realistic conditions matched to real environments.
Impact
Business outcomes
- 75% reduction in testing cycles, with improved test accuracy
- Non-technical users enabled to build sophisticated automated tests
- Launched September 2024; adopted by 20+ customers inside and outside Siemens
- Patent filed: 2024E15630 IN
- Siemens High Impact Professional Award, 2024
Reflection
What I would carry forward
The interaction model here did not start with this product. It was first proven in an earlier SIMIT training-simulation tool, where stakeholder response validated the approach. Rapid Tester was the tighter, faster-to-ship application of it — and the one that reached market first. Designing something once and having it become the vocabulary for a second product is the part of this work I would carry forward.
What came next
- Performance improvement
- Backend architecture optimisation
- Validating and incorporating user feedback after prioritisation
Strategic impact
Innovation Leadership
- Led design end-to-end as the lead designer, partnering with product, business, and development teams
- Aligned automation capabilities with engineering workflow needs
- Established design standards adopted across engineering tools
Strategic impact
Executive Stakeholder Management
- Presented innovation strategy to C-level executives
- Translated capabilities into business value propositions
- Influenced portfolio strategy and investment decisions
Strategic impact
Intellectual Property Leadership
- Pioneered user-centered approach to automation testing
- Established design patterns for low-code engineering UX
- Created IP foundation for competitive differentiation (Patent 2024E15630 IN)
- Siemens High Impact Professional Award, 2024.
Next case study
Energy Asset Monitoring Platform
Get in touch