ServicesStage 02 · Interpret02 / 11

Data Engineering and Analytical Foundations

Build a dependable foundation for analysis

Data foundationsources → ingest → store → use
SOURCESERPERP: Orders, assets, work orders · nightly batchHistorianHistorian: Process tags · 1 sSensorsSensors: Vibration, current · 25.6 kHz burstsDocumentsDocuments: Reports, logs · on changeINGESTCollectCollect: Batch and streaming connectorsValidateValidate: Units, timestamps, gaps, duplicatesLabelLabel: Asset identity, lineage, metadataSTORETime seriesTime series: Measurements by asset and tagAnalyticalAnalytical: Reconciled business and machine recordsUSEModelsModels: Training and scoringApplicationsApplications: Dashboards, decision supportReportsReports: Traceable to source
flow4 sources · 2 stores · 3 uses

Data engineering

Analytical performance depends on the availability, consistency and meaning of the underlying data. MNC Analytics helps organisations organise fragmented information and establish the data flows required for reliable analysis and deployment.

We work across enterprise systems, operational platforms and connected devices, addressing how information is collected, transformed, stored and made available to users and applications.

What we offer

8

  1. 01

    Data inventories, profiling and readiness assessments

  2. 02

    Data cleaning, reconciliation, labelling and quality controls

  3. 03

    Batch and streaming data pipelines

  4. 04

    Database, time-series and analytical storage architecture

  5. 05

    Integration of business records, machine data and sensor measurements

  6. 06

    Common definitions, metadata, lineage and asset identification

  7. 07

    Management of timestamps, measurement units, missing values and changing equipment configurations

  8. 08

    Access, retention and reproducibility arrangements appropriate to the intended use

What this enables

A consistent and traceable data foundation that supports analysis, reduces avoidable preparation work and makes results easier to investigate and maintain.

Related work

All work
01Energy and utilitiesFrance

Forecasting heat and electricity demand for a district utility

Hourly forecasts with calibrated uncertainty bands replaced spreadsheet estimates, so the dispatch team now commits plant and purchases balancing energy against a stated confidence level.

  • -52%mean absolute forecast error
  • 89%observed coverage of the 90% interval
  • 48 hhourly horizon, retrained weekly
Read more
Mean absolute error by month, manual versus modelManualModel
051015OctNovDecJanFebMarOct Manual: 8.10 MWhOct Model: 4.20 MWhNov Manual: 10.4 MWhNov Model: 4.90 MWhDec Manual: 12.9 MWhDec Model: 6.10 MWhJan Manual: 13.6 MWhJan Model: 6.30 MWhFeb Manual: 11.8 MWhFeb Model: 5.60 MWhMar Manual: 9.20 MWhMar Model: 4.40 MWh
MWhMar · 4.40 MWh
02Buildings and infrastructureUnited Arab Emirates

Energy performance across a commercial building portfolio

Metering, building management and weather data were unified for 27 buildings, and calibrated building models now rank retrofit options by verified saving rather than by estimate.

  • 27buildings on one data foundation
  • 5 of 6retrofits within modelled saving range
  • -11%portfolio energy intensity, year one
Read more
Modelled versus measured saving, first retrofit roundModelledMeasured
05101520B1B2B3B4B5B6B1 Modelled: 14 %B1 Measured: 13 %B2 Modelled: 9 %B2 Measured: 10 %B3 Modelled: 18 %B3 Measured: 16 %B4 Modelled: 7 %B4 Measured: 8 %B5 Modelled: 12 %B5 Measured: 11 %B6 Modelled: 10 %B6 Measured: 6 %
%B6 · 6 %
Next service · 03Advanced Applied Mathematics

Contact

Tell us about the problem you need to solve

You may have a defined project, an underperforming process or a technical question that needs investigation. We can help assess the challenge, identify the information required and define an appropriate next step.

Discuss your projecthq@mncanalytics.com