Physical
Conveyors, sorters, scanners, print-and-apply, scales, cameras, motors, photo-eyes, e-stops, workstations, limited mobile equipment, and people moving product.
TARGET / SMART WAREHOUSE
Designing Shared Understanding Across Physical and Digital Operations
Warehouse automation generated enormous amounts of machine data. The harder problem was helping people understand what mattered across the building.
Target’s automation environment connected machinery, HMI/SCADA, dashboards, vendor systems, reports, floor activity, maintenance, and planning across more than 50 sites.
As the sole UX designer on the core Jetson product team, I worked across Operations, Engineering, Control Center, site leadership, and technical partners to investigate where those systems stopped creating shared understanding.
How could one operational system help different roles understand what is happening, what it affects, and what should happen next?Explore the Smart Warehouse prototype →
One warehouse. Many representations of it.
The physical operation was continuous. The software was not.
Each source could be correct while still providing only part of the picture. A machine reported a state; Operations interpreted its effect; Engineering investigated a cause; Leadership considered a different time horizon.
Shared data did not automatically create shared understanding.
Automation controlsJetsonVendor applicationsOperational dashboardsReportsSlack / messagesExcelSplunkTraining systemsSupport proceduresPhysical floor checks
The product sat between four systems at once.
Technical
IT, OT, PLC/controls, HMI/SCADA, telemetry, network dependencies, vendor software, and proprietary technology boundaries.
Organizational
Operations, Senior Site Engineers, Maintenance, Engineering & Facilities, Control Center, site leadership, HQ, vendors, and support groups.
Decisions
Operations adjusted work. Engineering judged equipment intervention. Leadership assessed plan, productivity, cost, and risk.
Role-specific question
Where is work slowing, and what should change?
Flow, accumulation, labor, priorities, and downstream consequences shape the operational decision.
Decision emphasis
Adjust labor, plan, or release while there is still room to act; then recheck the effect.
Explore Material FlowOne shared reality. Different decisions.
Site requests were starting points, not requirements.
“We need First Pass Yield.” “Can I have another site’s dashboard?” “This alarm should text someone.”
I investigated the decision behind the request, current behavior, other systems consulted, and whether the need recurred across sites. Engineering joined when feasibility depended on equipment, controls, telemetry, networks, or vendor architecture.
- Local request / workaround
- UX investigation
- Cross-site comparison
- Technical validation where needed
- Reusable problem definition
- Design / prototype / pilot
- Learning feeds back into the system
Local workarounds were evidence of unmet needs, not specifications to copy.
Seven operational areas. One connected concept.
These modules are represented in the current Smart Warehouse prototype.
- Dashboard
- What is happening now?Whole-building state, developing conditions, major signals, and orientation.
- Material Flow
- Where is work moving, and where are consequences forming?Site Traffic and Site Plan; building → area → section → device.
- Smart Maintenance
- What needs intervention, and when can the operation tolerate it?Equipment state, repairs, planned work, escalation, performance, and lifecycle context.
- My Hub
- What do I need to solve, learn, follow, or reuse?Personal work, people, notifications, training, community, solutions, and site knowledge.
- Reporting
- How did the operation perform over a defined period?Shift handoff, site-period reviews, OT/IT summaries, OEE, and analytical tools.
- Historical
- What happened, what did we try, and what did we learn?Investigations, experiments, improvements, recurring patterns, and investment analysis.
- Settings
- What makes this site, user, or worker group different?Site configuration, integrations, equipment, alerting, roles, and worker-group settings.
CONCEPT · This prototype does not recreate every Target site or production screen. It connects recurring problems, delivered work, and future directions. Smart Maintenance was an engineering-led initiative; my contribution focused on UX, interpretation, and operational context.
The consequence reaches Repack
Follow the problem that looks familiar.
Material FlowFrom Local Metrics to Whole-Building FlowA team could hit its own target while creating a bottleneck somewhere else.Cross-site research connected yield, rejects, recirculation, gap timing, labor, equipment, and downstream congestion.Explore Material Flow →Human-Generated Signals / SICKWhen the Machine Can’t See the ProblemWorkers could recognize problems automation could not detect.A physical three-state signal and digital visibility turned human observation into shared operational state.Explore Human-Generated Signals / SICK →Equipment Signals → InterventionFrom Equipment Signals to Coordinated InterventionAn alarm could report a change without explaining whether someone needed to act.Alerting, diagnostics, role relevance, escalation, and maintenance context connected machine condition to response.Explore Equipment Signals → Intervention →
Standardize the repeatable pattern. Preserve the meaningful difference.
Current: copy and customize
Copied site implementations diverged, repeating maintenance and limiting consistent enterprise reuse.
Proposed: shared foundation
Enterprise source of truth → site-specific configuration.
~90% reusable foundation · ~10% meaningful variation
Engineering feasibility research I initiated pointed toward this model. Installed equipment, telemetry, vendor systems, thresholds, terminology, and controls could still vary.
PROJECTED ARCHITECTURE · NOT ACHIEVED ADOPTION
Explore the scaling story →What UX could influence. What required broader change.
UX could influence
Shared models, clearer state, role-aware interpretation, workflow structure, alert relevance, training, interaction patterns, cross-site synthesis, and reusable product concepts.
Broader change was required
Vendor access, proprietary systems, network architecture, uneven data, equipment differences, organizational ownership, enterprise repositories, and reliable predictive modeling.
The hardest problems were rarely contained inside one screen, one system, or one team.
Design the connective tissue.
Smart Warehouse reinforced a pattern throughout my career: complex products become difficult when people must reconstruct the real situation across disconnected systems.
State → context → consequence → ownership → decision