28  Big Data Case Studies

analytics-ml
big
data
case

28.1 Start With the Story

Picture an IoT team using the ideas in Big Data Case Studies during a live operations review. A device has produced messy evidence, an analytic step is about to change an alert or control decision, and someone has to explain why the result should be trusted.

Read this page as that path from sensor evidence to accountable action. Start with what the system observes, keep the model or data treatment visible, and finish with the check that would convince an operator, maintainer, or auditor to act.

28.2 Case Studies Start With Decisions

An IoT big-data case study is useful only when it connects a decision to the evidence path that supports it. “A city has millions of readings” is not enough. The case study must state the operational decision, the sensors involved, the sampling rates, the data contracts, the processing path, the retention rule, and the measured outcome. Otherwise it is only a volume story.

Good case studies also separate raw data from decision evidence. A traffic platform may keep raw video locally or in an archive, but dashboards often need counts, speeds, incident flags, confidence scores, and short event clips. A maintenance platform may keep raw vibration bursts for review, but streaming alerts often depend on features such as RMS, spectral energy, temperature trend, load state, and model version. The architecture is credible when the retained evidence can be replayed, audited, and tied back to a decision.

The first question is not “which big-data tool is fashionable?” It is “which decision is being made, which evidence proves it, and what must be retained if the decision is challenged later?”

Single laptop versus distributed system comparison for IoT big-data processing.
Case-study scale changes architecture: once the evidence path exceeds one machine’s memory, runtime, or failure tolerance, the design must distribute processing while preserving replayable decision evidence.

Decision

Name the action: reroute traffic, dispatch maintenance, change a cooling setpoint, or quarantine a bad sensor.

Evidence

Define the readings, features, quality flags, model versions, and raw windows needed to defend the action.

Architecture

Map evidence through MQTT, Kafka, Flink, Spark, object storage, data lake tables, and serving stores.

Outcome

Measure latency, cost, avoided downtime, energy reduction, safety response, or analyst time saved.

Case Type
Decision
Evidence Path
Main Risk
Smart traffic
Change signal timing, route crews, or publish congestion alerts.
Loops, camera metadata, edge object counts, Kafka topics, stream windows, and dashboard aggregates.
Raw video privacy, late events, incorrect time alignment, and false incident detections.
Industrial maintenance
Create a work order before a motor, pump, or conveyor fails.
Gateway features, vibration windows, model id, anomaly score, CMMS ticket, and post-maintenance result.
Concept drift, uncalibrated sensors, false alarms, and missing raw evidence for review.
Cold chain logistics
Accept, inspect, reroute, or reject a shipment based on temperature exposure.
Device telemetry, gateway buffer, time-series store, excursion rules, proof-of-custody records, and reports.
Connectivity gaps, clock drift, tamper evidence, and unclear responsibility at handoff.
Building energy
Adjust HVAC schedules and setpoints while maintaining comfort.
Occupancy counts, CO2, temperature, weather, BACnet or MQTT streams, forecasts, and control logs.
Comfort regressions, actuator limits, model bias, and uncontrolled manual overrides.

Overview Knowledge Check

28.3 Build the Case Ledger First

A practical case ledger converts a scenario into rates, payloads, storage tiers, and query paths. The numbers below are a representative smart-city workload, not a claim about one named city. The point is the calculation method: quantify the streams, separate metadata from heavy media, and make storage and processing choices from measured demand.

The same ledger also prevents architectural overreach. Parking sensors and air-quality stations may fit batch summaries. Traffic-loop counts may need sub-minute stream processing. Camera-derived events may need edge inference and privacy controls. A single relational database or a single data lake table is unlikely to serve all access patterns cleanly.

Worked case: representative smart-city platform
parking occupancy:
800 sensors * 2,880 readings/day * 80 bytes = 184,320,000 bytes/day = 0.184 GB/day

air quality:
200 stations * 1,440 readings/day * 220 bytes = 63,360,000 bytes/day = 0.063 GB/day

streetlight status:
15,000 lights * 1,440 readings/day * 90 bytes = 1,944,000,000 bytes/day = 1.944 GB/day

traffic-loop counts:
3,000 loops * 86,400 readings/day * 120 bytes = 31,104,000,000 bytes/day = 31.104 GB/day

camera event metadata:
600 cameras * 172,800 events/day * 400 bytes = 41,472,000,000 bytes/day = 41.472 GB/day

event clips selected by edge inference:
600 cameras * 200 clips/day * 2 MB = 240,000 MB/day = 240 GB/day

daily total:
0.184 + 0.063 + 1.944 + 31.104 + 41.472 + 240 = 314.767 GB/day

design reading:
The platform is not dominated by parking or air-quality records. It is dominated
by selected camera clips and traffic streams. That pushes the architecture toward
edge inference, hot summaries, lifecycle-managed object storage, and strict access
control for media.

Case-study math should change the design. If one stream contributes 76 percent of daily volume, retention, privacy, compression, and query strategy should focus there first.

Stream
Ingestion
Storage and Query
Operational Evidence
Parking and air quality
MQTT broker to Kafka or cloud IoT hub with device identity and schema validation.
Parquet tables in Delta Lake or Apache Iceberg, plus hourly aggregates in ClickHouse or Druid.
Device heartbeat, missing-window count, calibration date, schema id, and late-event count.
Traffic-loop counts
Kafka partitions by corridor or region, with Flink or Spark Structured Streaming windows.
Hot stream aggregates for dashboards and compacted history for seasonal analysis.
Partition lag, window completeness, watermark delay, and dashboard latency.
Camera events
Edge object detection on the gateway, then metadata and selected clips to cloud storage.
Object storage with lifecycle rules; metadata indexed in OpenSearch, Pinot, Druid, or ClickHouse.
Model version, confidence, clip id, access log, redaction state, and retention expiry.
Cross-domain analytics
Curated lake tables joined by time, location, sensor type, and event id.
Notebook, SQL, and dashboard workloads over governed curated zones.
Lineage, table version, query scan bytes, owner, and business metric impact.

A second case shows why big-data systems do not always need the same stack. In cold-chain logistics, the main decision is whether a shipment stayed within temperature limits. A gateway may buffer BLE or cellular sensor readings during connectivity gaps, then upload signed time-series records. TimescaleDB, InfluxDB, or a cloud time-series service can serve recent queries, while object storage preserves signed audit bundles. The critical evidence is not petabytes of volume; it is timestamp integrity, custody handoff, clock sync, tamper state, and rule evaluation.

Practitioner Knowledge Check

28.4 Validate Economics and Evidence

Case studies become engineering assets when their numbers can be checked. Daily volume should follow from device count, sample rate, and payload size. Query latency should be measured from real dashboards or replay tests. Storage cost should separate hot, warm, and archive data. Outcomes should name the baseline and the measurement window: fewer truck rolls, shorter incident response, reduced energy use, lower spoilage, or faster analyst workflow.

Cloud prices, storage classes, and managed-service charges change, so procurement should always refresh current pricing. The calculation pattern is still stable: estimate retained GB by tier, multiply by the contract price for that tier, add compute, network egress, observability, and operations labor, then compare the result to the operational value.

Worked example: tiered storage for the smart-city case
daily generated evidence: 314.767 GB/day
hot retention target: 7 days
warm retention target: 90 days total
archive policy after 90 days: keep only 20 percent of records and clips

example storage prices from a contract, not a universal quote:
hot object storage: $0.023 per GB-month
warm object storage: $0.012 per GB-month
archive object storage: $0.004 per GB-month

all data hot for 90 days:
314.767 * 90 = 28,329 GB
28,329 * $0.023 = $651.57/month

tiered first 90 days:
hot: 314.767 * 7 = 2,203 GB; 2,203 * $0.023 = $50.67/month
warm: 314.767 * 83 = 26,126 GB; 26,126 * $0.012 = $313.51/month
first-90-day tiered total = $364.18/month

archive after 90 days for a seven-year record:
archive days = (7 * 365) - 90 = 2,465 days
archive retained volume = 314.767 * 2,465 * 0.20 = 155,178 GB
archive monthly storage = 155,178 * $0.004 = $620.71/month

design reading:
The storage bill is not the only cost, but the calculation proves why lifecycle
policy and archive sampling are architecture controls, not cleanup tasks. It
also shows which assumptions must be reviewed when actual access patterns arrive.

Replay Evidence

Keep topic offsets, schema ids, table versions, raw object paths, and job versions so a result can be rebuilt.

Privacy Evidence

Record access logs, redaction state, retention expiry, consent or policy basis, and review decisions for sensitive data.

Cost Evidence

Track retained GB by tier, query scan bytes, compute hours, network egress, small-file counts, and owner tags.

Outcome Evidence

Compare decisions against a baseline: downtime avoided, incidents detected, trips reduced, energy saved, or spoilage prevented.

The under-the-hood lesson is that a case study must be falsifiable. If the dashboard claims sub-second traffic visibility, logs should show event-time delay, partition lag, query latency, and missing-window rate. If the maintenance system claims avoided downtime, the work orders should connect model alerts to inspections, repairs, and avoided failures. If the architecture claims a lower cost, the cost report should show which lifecycle rule, partition layout, and materialized view produced the reduction.

Under-the-Hood Knowledge Check

28.5 Summary

  • IoT big-data case studies should start with a decision, not with a tool list or a headline sensor count.
  • A case ledger connects sensor rates, payload sizes, evidence contracts, ingestion paths, storage tiers, query systems, operating metrics, and measured outcomes.
  • Representative smart-city workloads often become media, privacy, and lifecycle problems before they become pure database problems.
  • Industrial, logistics, and building-energy cases may have lower total volume but higher evidence-integrity requirements.
  • Strong case studies are falsifiable: their data rates, replay paths, cost assumptions, access controls, and outcome claims can be checked.

28.6 Key Takeaway

A useful IoT big-data case study is an auditable architecture ledger. It names the decision, quantifies the data streams, separates raw data from decision evidence, chooses storage and processing systems from access patterns, and preserves enough operational evidence to replay, govern, and measure the result.

28.7 Common Pitfalls

  • Treating a high sensor count as proof that a big-data platform is needed, without calculating rates, payload sizes, or query workload.
  • Keeping every raw byte in the hottest tier instead of matching retention to replay, audit, privacy, and access frequency.
  • Selecting Kafka, Spark, Flink, or a lakehouse table format before defining the evidence contract.
  • Reporting an outcome without a baseline, measurement window, or link from data-driven decision to operational result.

28.8 See Also

If you want to… Read this
Review the big-data concepts behind the cases Big Data Fundamentals
Compare stream, batch, and storage technologies Big Data Technologies
Design ingestion and transformation pipelines Big Data Pipelines
Reduce high-volume streams before cloud upload Edge Processing for Big Data
Operate replay, retention, freshness, and cost controls Big Data Operations
Connect case-study choices to cloud services Cloud Data Platforms and Services