Semantics-Driven Middleware Layer
for Building Operation Analysis in
Large-Scale Environments
Adam Kučera
Outline
 Introduction & Motivation – Facility Management Systems
 Problem – BMS Data Analysis
 Methods & Areas of Research
 Results
 Conclusions
Introduction
Facility Management Systems
Facility Management
 According to IFMA (International Facility management
association): „a profession that encompasses
multiple disciplines to ensure functionality of the
built environment by integrating people, place,
process and technology“
 FM ensures tasks, which are not part of organization‘s
„core business“
CAFM(Computer-Aided Facility
Management)
 CAFM software supports:
 Space management
 Maintenance
 Energy management
 Provides advanced analytical
tools
BIM (Building Infrastructure Modelling)
 Database of building constructions and devices
BMS (Building Management System)
 Monitors and controls building automation systems
 MU has large BMS (40 buildings, 1000 devices,...)
BMS (Building Management System)
Motivation
 Facility manager should be able to query the BMS system
in similar manner to those examples:
 Show me which rooms on the second floor of A11 building
had running AC units during last 8 weekends.
 Tomorrow morning, I want to receive report about
electricity consumption in 5 minute intervals for those 4
buildings since now.
 I want to know which devices influence temperature in
office of Mr./Mrs. XY.
 For all buildings at University Campus, compare electricity
consumption per square meter.
Problem
Issues of Building Operation Analysis
Issues of BMS
 Inaccessible data
 Missing semantics
 Inflexible built-in analytical features
=>
 Advanced analytical tools are unavailable for large-scale
environments
 Integration of BMS, BIM and CAFM does not exist
Two Types of Users
 BMS contain precise and detailed data about building
operation
 Those data are not easily accessible
 Two kinds of people:
 Knows how to analyze data but can‘t get them (Facility
managers)
 Knows how to get the data but can‘t analyze them (BMS
operators)
Methods & Areas of Research
StepsTowards Flexible And Efficient Analysis
Methods and Areas of Research
 Middleware
 Data access
 Data semantics
 BMS specific operators
 Query library
 Front-end applications
Results
Existing Applications &Tools,Work in Progress
 Exposing data
 Technology data mart
 BMS API
 Integrating BMS&BIM
 Ontology repository
 Analysis & UI
 CEP engine
 Archive data browser
 Machine learning methods
Results
Conclusions
Benefits of Proposed Solution
Conclusion
 The main goal:
 Middleware for FM data processing
 Developers will focus on:
 Analytical methods
 Convenient user interfaces
 Facility managers will be provided with:
 Direct querying of the BMS
 Flexible reports
 Advanced analytical tools
 Incorporation of BMS data into CAFM
Thank You for your attention.
Questions?