Business Challenge story
Across the building industry, facilities management has become increasingly data-centric, with an enormous range of sensors constantly producing data from such sources as lighting; occupancy; security; heating, ventilation and air conditioning (HVAC); and electrical systems, to name only a few. However, even in today’s digital age, building management is generally performed on proprietary system-by-system and localized building-by-building bases. Older building-automation systems suffer from limited or no integration capabilities, inadequate analytics tools, proprietary data interfaces, and a cacophony of unintegrated software applications and user interfaces. Although device information can help facilities managers improve each individual system’s efficiency, there is often little visibility into synergies and connections across multiple systems.
The facilities-management services team at Carnegie Mellon University (CMU) instinctively knew that it could find significant energy and operational savings if it had better information about the operating condition of its assets; however, the university’s existing building-automation systems were not designed for analytics and were not integrated to manage a portfolio of facilities. This meant that although building operators could view data from separate assets to see if they were functioning within parameters - for example, keeping the ambient temperature of a building or room at a certain level - operators might miss the fact that to maintain that temperature, both the heating and air conditioning systems sometimes ran simultaneously. Often, experienced facilities managers would spot and fix such anomalies based on their in-depth knowledge of the systems, but their insights could not be captured by or integrated into the building-management system, so problems would continue unchecked, wasting energy and reducing the life span of equipment.
As a world-class research institution and one of the leading building sciences and engineering universities in the United States, CMU knew that it needed a visionary, interconnected building-management system across its campus. Although it was feasible to build such a system internally, designing and implementing a smart campus from the ground up could be time-consuming, and implementing and maintaining the supporting IT infrastructure could be costly in terms of both capital and operating expenditures. Further, such capital IT projects at CMU compete with similarly priced capital building improvements. The university wanted not only a best-in-class integrated facility-portfolio management system, but also one that would help it reduce operating expenses by reaping energy cost savings as quickly as possible.
This idea led the associate vice president for facilities management to seek an industry-leading organization to help the university make its Smarter Campus vision a reality. That organization would be challenged to implement a world-class, scalable solution that would essentially pay for itself by reducing energy and maintenance costs across CMU’s main campus and that could scale from a few buildings to multiple campuses with limited capital investment. CMU reached out to IBM because it already used IBM Maximo Asset Management software to manage calendar-based maintenance and automatically generate work orders for its facilities portfolio. After discussions about creating a smart campus, IBM and CMU collaborated to apply the concept of analytical, interconnected and automated facilities management to a portfolio of campus buildings. The university’s goals were not only to reduce energy expenditures through improved building efficiency, but also to increase the productivity of operations staff and harvest actionable, repeatable insights for future application in the field of facilities management.
Building-automation systems are excellent sources of raw data, but when that data is self-contained in separate, proprietary systems, it defies comprehensive analysis and yields little functional insight. Carnegie Mellon University is unlocking deeper intelligence from building-asset data by implementing a unified facilities-management solution.
Because the solution can monitor and make sense of data from both a building’s individual components and its system-level conditions as an integrated whole, it can quickly find the root cause of problems. Cloud-based analytics software from IBM Business Partner SkyFoundry not only helps identify anomalies, but also provides likely solutions and work-order prioritization based on an issue’s criticality and cost impact. For example, a building operator will receive not only an alert about an overheated lecture hall, but also a diagnostic report that directs him to the mechanical system so that he can address the vent that is stuck open and forcing the building’s heating system to run constantly. Maintenance teams can avoid time-consuming manual diagnostics, meet customer needs faster and improve equipment longevity.
A sales team and consultants from IBM Global Business Services - Business Consulting Services is working with CMU to design a solution that not only meets the university’s stated energy-savings and fiscal needs, but also one that it could showcase and use to educate its own students and other universities about the value of modern building-data management and analytics. The IBM Smarter Buildings solution will use IBM TRIRIGA software and SkySpark analytics engine software from IBM Business Partner SkyFoundry as the foundation of a new cloud-based building-management system to help increase the operational, financial and environmental performance of the university’s facilities. IBM will host the building management and analytics solution as a subscription-based software as a service (SaaS) offering on the SoftLayer platform, which provides the cost flexibility and scalability that CMU is seeking. Because the university will pay for the solution as a monthly subscription and will not need to implement or maintain its own hardware or software, the state-of-the-art solution will fall under the university’s facilities operating expenses rather than under the capital expenses budget.
The university currently has a 3-year solution subscription to the SoftLayer platform-hosted solution, and CMU will apply it to 10 buildings for the first six months and then expand it to 36 buildings, covering 4.5 million square feet. During solution deployment, Business Consulting Services teams will provide analytics as a service (AaaS) consulting to help CMU adopt the solution, interpret analytics and optimize the solution’s value potential. The solution pulls data from building information systems, then aggregates, rationalizes and passes it to the SkySpark analytics engine for analysis. Applied analytics focus on desired outcomes - for example, efficient operation of a building’s HVAC system - rather than on individual set points or equipment. The analytics engine analyzes and compares variables and data both within and across systems to spot anomalies and provide root-cause analysis to help pinpoint both the location of and reason for the problem. The SkySpark software sends an alert through the TRIRIGA software, which coordinates individual and system alerts with aggregated building data to create specific, prioritized service requests that it then passes to the Maximo Asset Management software. The integrated work-management system reports when corrective maintenance has occurred and monetizes it to record the savings.
In the past, operations personnel might have received a complaint that a lecture hall was overheated, and they would subsequently check the signals from that classroom and input this service alert into the Maximo software, which would generate a work order sending a technician to the building. Once there, the technician would have to investigate further if the cause was not readily apparent, going from floor to floor to see whether a vent was stuck in an open or closed position, causing the heating system to continuously run and overheating the lecture hall in question. Now, the building-analytics solution will continuously and automatically analyze data from multiple sensors within the building. In this case, even before the facilities managers receive a complaint, the solution will create a preventive alert about an existing anomaly that would overheat the lecture hall if left unattended. Additional intelligence comes from comparisons, when signals indicate that a classroom on a different floor cannot be kept at the required minimum temperature.
By comparing the data from these two alerts, the TRIRIGA software can conclude that the heating system has deeper problems. Now, when the Maximo software generates the work order, it will contain detailed information that spans systems and spaces to help the technician quickly fix the problem instead of spending time researching its cause or addressing just the symptom.
Projected to reduce energy costs by at least 10 percent and mechanical operations costs by at least 8 percent, or USD2 million annually, which savings the university can use to recoup its investment in approximately 24 months; Anticipated to encourage innovation with cloud-based deployment because the university can add or change the software solutions without further IT investment; Expected to help the university move from calendar-based to proactive, condition-based facilities management