Results from testing of a “cloud based” automated ...

Results from testing of a “cloud based” automated fault detection and diagnosis tool for AHU’s Ken Bruton

Daniel Coakley

I2E2 Researcher

I2E2

Department of Civil &

Peter O Donovan

Researcher Informatics

I2E2 Researcher Informatics

Marcus M. Keane

D. T. J. O'Sullivan

Lecturer &

Lecturer &

Principal

Principal

Investigator

Investigator

Environmental

Research Unit for

Engineering,

Sustainable

Research Unit for

University College

Engineering,

Sustainable

Research Unit for

Civil &

Cork, Ireland

National

Engineering,

Sustainable

Environmental

University of

National

Engineering,

Engineering,

Ireland, Galway

University of

National

University

Ireland, Galway

University of

College Cork,

Ireland, Galway

Ireland

Abstract Heating Ventilation and Air Conditioning (HVAC) system energy consumption, on average, accounts for 40% of an industrial site’s total energy consumption. Studies have demonstrated that continuous commissioning of building systems for optimum efficiency can yield savings of an average of over 20% of total energy cost. Automated Fault Detection and Diagnosis (AFDD) can be used to assist the commissioning process at multiple stages. This paper outlines the development of an AFDD tool for AHU’s using expert rules. It outlines the results of the alpha testing phase of the tool on 18 AHU’s across four commercial & industrial sites with over €104,000 annual energy savings detected by the AFDD tool.

1. Introduction The contribution of buildings towards total worldwide energy consumption in developed countries is between 20% and 40%. This is expected to rise by an average rate of 1.5% per annum over the next 20 years [1]. Heating Ventilation and Air Conditioning (HVAC), energy consumption, and more specifically the operation of the Air Handling Units (AHU’s), accounts on average for 40% of an industrial site’s total energy

Informatics

Department of

consumption [2] due primarily to the stringent cleanliness requirements that many of the industrial processes (clean-rooms, pharmaceutical production, etc.) require to be in compliance with international standards [3]. Buildings rarely perform as well in practice as anticipated during design due to improper equipment selection or installation, lack of commissioning, or improper maintenance [4] to cite but a few reasons. Studies have indicated that 20 – 30% energy savings are achievable by recommissioning HVAC systems, and more specifically AHU operations, to rectify faulty operation [5]. Studies have also demonstrated, using a sample set of over 80 buildings, that continuous commissioning of building systems for peak efficiency can yield savings of an average of over 20% of total energy cost [4]. By coupling the re-commissioning and ongoing commissioning of HVAC systems into one demonstration study, savings of 44% of electricity consumption and 78% of gas consumption over a ten year period have been proven by the International Energy Agency (IEA) Annex 47, DABO Case Study[6], [7]. Building Energy Management Systems (BEMS) are typically the repository of HVAC system and more specifically AHU operational data and are now

Figure 1: Example off a typically undetected fault in a heating coil control valve A typical example is the hot commonly installed in commercial and industrial h water energy wasted by buildings [8]. Typically the cost of instaalling a BEMS is a passing heating coil contrrol valve, as illustrated in the biggest constraint to doing so. Appproximately 45 – Figure 1 by the 5 oC rise in teemperature across the coils 75% of the associated cost is due to wirring (materials & in the AHU when both coontrol valves are showing labour) [9]. However, as the cost off wireless sensor closed. This particular exampple typically goes unnoticed networks decreases and their robustness r and for long periods, as it is often possible for the AHU to maintain control of all set-pooints by using the cooling interoperability increases, so too will their application in building energy management systemss [9]. This will coil simultaneously to counteract the heating effect of the passing heating coil. serve to reduce the cost and expand thhe application of these systems still further. The availabbility of BEMS’s offers the potential for AFDD systems with w lower set-up 3. How can AFDD tools t improve AHU costs than cases where a BEMS is not present. maintenance and heence performance

2. AHU Maintenance AHU operations are typically supervised and maintained by either an onsite facilities team or an offsite third party contractor. The number of AHU’s in a typical HVAC system often outtnumbers those supervising and maintaining the system m by 20 to 1. This means that routine mechanical maintennance is typically carried out only when necessary due to an end user complaint, a machine breakdown or a breached alarm limit. The complexity of modern AHU control philosophies also commonly results in onsite personnel not having the required knowledge to root r cause issues without costly external consultancy. Buuilding operators are also typically overwhelmed by AHU A data as the overlying monitoring and control systems have had little effort put into consolidating the vast v quantities of information into a clear and coherent format [10]. BEMSs are typically used to supervise the performance of AHU’s in HVAC systems, s raising alarms when upper or lower limits of o operation are breached. However, they do not diagnose the root cause of these alarms, with few current BMS S systems having fault detection or diagnostic capabilities. Furthermore, when no alarm levels are breached, thhey do not detect underlying faults during what appearrs to be normal operation of these systems.

In order to ensure effectivee maintenance of an AHU, it must firstly be set up to operate effectively at the commissioning stage. As HV VAC systems grow more complex, so too will thee commissioning process required to ensure their inittial efficient operation. If current trends continue this movement towards more complex systems will result in a more expensive and lengthy commissioning proceess. Xiao & Wang (2009) support this hypothesis statiing that commissioning is labour intensive and that thee future will be that of an automated lifecycle commisssioning process embedded in the operation of the buillding management system [11]. Xiao & Wang (2009) also state that AFDD is a key means of achieving autom mated commissioning [11]. IEA Annex 47 [6] reviewedd the operation of 18 such tools, concluding that autom mation is still uncommon during the commissioning proocess, and that future tools should be developed that are easily embedded in existing operational practices.. Most commissioning activities a are currently performed manually, as labour-intensive, l once-off undefined processes upon com mpletion of an installation. For commissioning to be truuly successful, it must be embedded in the overall project p process, and must follow a defined procedure in order to ensure it is effective and repeatable. It was w for these reasons that the IEA launched Annex 40 4 [12], with a view to developing, validating, and documenting d tools for the

commissioning of buildings and theiir services. This study concluded that there are four ideal types of commissioning namely; 1) Initial commissioning – a firrst time process carried out on a newly construccted building; 2) Retro commissioning – a first time process on an existing building where no documented commissioning was previouslyy carried out; 3) Re-commissioning – a processs undertaken on a new or existing building to veerify or improve performance; 4) Continuous commissioning – undertaken continually to maintain, improove and optimise performance. t AHU’s in a By firstly re/retro commissioning the HVAC system, their performance can be b set to optimum levels. This would serve to return the AHU’s in the HVAC system to optimal “fault free”” operation. This could be achieved using an AFDD tool,, which identifies a number of areas which are not perform ming as intended. Once these items have been remedied,, the AFDD tool could be used as an ongoing commiissioning tool to ensure that any new faults in the systtem are detected effectively. Using this development model, curreent constraints in terms of maintenance resource availability a and knowledge base could be overcome. The T maintenance of large numbers of Air Handling Uniits (AHUs) by a relatively small number of resources would now no longer be an issue as an AFDD tool couuld generate a list of prioritised action items focusing effforts where most required. The level of knowledge requirred to ensure that the AHU’s in a HVAC system are opperating correctly could be reduced as an AFDD tool could provide a reasoning engine to detect and diagnose d issues automatically, only notifying the user when w an issue has been diagnosed. Lastly, the issues assocciated with nonoptimal control strategies or a lack of understanding u of their operation could be addressed ass an AFDD tool could monitor both the initial conntrol set-up and

subsequent operation of the coontrol algorithms, ensuring their robust and repeatable opperation.

4. AFDD Tool Develop pment The primary objectives off the AFDD tool developed as part of this project were the t development of a data access layer which had the fllexibility to work with any BMS type or age as well as the t ability to process BMS data into a generic form for poost processing. It must also have a busiiness layer which has the flexibility to work with any combination c of sensors and components found in typicaal AHUs to ensure cross company compatibility, whilee also having the ability to use already available measurrement without the need to install additional sensors to limit associated installation costs. Most critically howeveer, it must do all this while delivering a low number of false f positives/negatives in order to build confidence in thhe tool. The AFDD tool must be designed to have a user friendly graphical user interfaace (GUI) which allows the evaluation of on-going perfoormance of AHUs without in depth domain expertise. Thhis GUI must also facilitate the quantification and prioriitisation of the diagnosed faults to ensure return on investment is maximised through the repair of the mostt cost effective faults first. All of this also had to be achieved a with a rapid setup time per AHU to facilitate quick q set up and maximum scalability across an organisattion architecture utilised to develop the AFDD tool. The novelty of this AFDD D tool is in the generic way that it accesses data from f disparate building management systems allowinng the data to be analysed and the identified faults priooritised by the proprietary JAVA based business logicc. The simplified system architecture for this AFDD tool is described in Figure 2.

4.1 Data Access and Proccessing There are a number of issuues that cause difficulty in obtaining BMS data and preeparing and transferring it

Figgure 2: AFDD Tool System Architecture

for analysis. Firstly, there are many different BMS’s in operation in different companies each most likely procured from different vendors. Each BMS software has its own proprietary method of archiving data. These archiving methods cover a wide range from very primitive (e.g. one comma separated value file for each sensor/or component in which each line in the file corresponds to a daily dump of all of the values stored on the controller) to advanced (e.g. a normalised relational database). Secondly, getting robust data can be difficult due to the age of the BMS software which is seldom kept up to date. Due to this, there are often missing values, irregular timestamps and spurious outliers in the data that need to be corrected or removed from the dataset. Thirdly, measured data from AHU’s is often not archived on many installations due either to a lack of understanding of the benefit and the process to achieve this by the system owner, or due to the storage required to archive the quantity of AHU data monitored by a typical BMS. A generic data access tool was hence developed which could upload BMS data from the client server irrespective of the type of BMS software being utilised. Once accessed and retrieved via the data extraction tool, the data must be processed to ensure it is generic, irrespective of the system that it is integrating with. As there is no common naming convention used across all BMS systems and their associated control and monitoring equipment, each sensor/component id tag on the BMS has to be mapped to a particular sensor/component in a database. The naming convention for each recorded value is then standardised within both the database and the AFDD tool so that different AHU’s can easily be compared at a glance. Next, all of the values are converted to a single format for clarity and ease of use across all sites and AHUs. Finally, the processed measured data for each site is stored in a remotely hosted relational database management system (RDBMS) using MySQL. Static information about each AHU is also required in addition to the measured BMS data. This static data is acquired during the initial site survey using a configuration page where data is input manually at set up stage and uploaded to the MySQL database.

4.2 Fault Detection Business Layer Correct identification of the mode of operation of an AHU is critical to identifying how each component should operate, and therefore, critical to performing AFDD. Approaches to date have focused on determining the operating mode of the AHU by identifying the actual operating mode from the outside air damper position, cooling and heating coil valve positions [13], [14], [15]. The tool described in this

paper incorporates this approach but with some additions to increase effectiveness. The business layer of the AFDD tool was developed using the APAR rule set [13] as a starting point. Significant modifications and improvements have been made to this rule set over the course of this project namely; • Expansion of the 28 APAR rules to over 60 in number, to account for other possible configurations of components and sensors: such as reheat coils and off coil temperature sensors; • Expansion to include capability to detect faults on fresh air and not just return air AHU’s • Addition of virtual readings calculated using existing data: • Development of a methodology to determine the “ideal” mode of operation • Development of an improved statistically based error threshold analysis Once data is extracted from the BMS, a client side JAVA application performs mode checks, calculates virtual values, applies business layer rules, and stores consequent results in a MySQL database. A key issue identified in the early phase of this project was the use of a single, heuristically defined error threshold value that was applied to all of the rules. The purpose of this error threshold is to reduce false positives: where a fault is identified by the tool but none is present in the AHU. This error threshold is intended to account for all of the inaccuracies in measurements due to sensor error, sensor location, whether the measurement is taken using an averaging or point sensor, and the dynamic nature of AHU operation. A major issue with previous approaches is that the actual error associated with each rule is dependent on the number and type of measurements used in a particular rule. For example, a rule using just one measurement from an averaging temperature sensor should have a lower error threshold than a rule using 3 measurements from a variety of sensors. The approach used in this tool allows the user to define an individual error threshold associated with each sensor. The error thresholds for each sensor measurement used in a fault check are then combined to yield a rule-specific error threshold.

4.3 Graphical User Interface A GUI has been developed for the AFDD tool. The GUI has been built in Microsoft Excel using Visual

Basic for Applications (VBA) (Figure 3). This tool is very much a development tool which will be transitioned to a more scalable solution later in the project. The GUI allows the user to set up new AHUs in the AFDD tool, modify set up parameters and most importantly to visualise results. The GUI automatically generates a schematic of the AHU based on the components and sensors that are present within the unit. Users can select any particular instance of time to be presented on this schematic. This feature allows the user to quickly view a representation of the AHU under any past operating conditions allowing fault instances to be analysed in more detail. All identified faults are listed by frequency of occurrence in the analysed period. Once a particular fault is selected, the GUI displays: •

The measured and virtual data on the schematic for the most recent hour in which the fault occurs A textual description of the currently displayed fault; A graphical trend of the past 48 hours of values relevant to that fault; A list of possible diagnoses for the currently displayed fault;

• • •

This allows for rapid viewing of individual faults, but still requires partial user input to root cause the fault at this stage of development. A number of potential methods to diagnose root causes are currently under Site

No. AHUs

1

2

2

4

3

9

4

3

Type

Constant volume Constant volume Variable Volume Variable Volume

Design airflow [m3/s] 14

development including the use of forward/ backboard chaining algorithms and case based reasoning and will be tested as part of the beta testing phase of the project.

5. AFDD Tool Alpha Testing 5.1 Design of Test Study The Innovation for Ireland’s Energy Efficiency (i2e2) Technology Centre is an Irish government sponsored initiative. The i2e2 research focus is on energy efficiency improvements in manufacturing processes and supporting systems.. The main objective of the current tranche of research is to provide an automated FDD tool that has been extensively tested on a range of different AHU’s across a number of disparate industrial sites Several hundred AHU’s were available for selection as part of this project. A number of characteristics were taken into consideration in order to select a number of representative units for the purpose of developing and validating the tool. As such, AHU’s were selected with different component and sensor layouts in order to ensure that the AFDD tool can be applied effectively and comprehensively. They also had varying levels of instrumentation in order to alleviate concerns regarding the level of instrumentation needed to perform FDD effectively. Table 1, in which the industry partners are anonymous, details the pilot AHU characteristics.

Type of zone(s) supplied

Operating hours per annum

BMS Software

Frequency of logged data

Office & canteen

8760

Trend

15 minutes

20

Production area

8760

Tridium

15 minutes

13

Manufacturing Floor Commercial office space

6240

Cylon

15 minutes

6240

Schneider

15 minutes

8

Table 1: Characteristics of BMS and data acquisition methods on each site

5.2 AFDD Tool Test Path In order to test the effectiveness and robustness of the AFDD tool developed as part of this project, it was decided to undertake initial alpha tests on four pilot sites supported and monitored by project researchers. This involved getting data automatically form each site to one remote database. All analysis and GUI interpretation was then undertaken in the research lab by skilled researchers who had developed the tool. This

allowed the GUI development to be kept to a minimum as well as minimizing the work necessary to make the output from the tool understandable to someone who was not a member of the development team. This meant that results from the AFDD tool could be more quickly brought to the end user. After a successful period of alpha testing, it is envisaged that a period of beta testing will

annum. These faults have been quantified based on the thermodynamic properties of the air traveling through the AHU and verified by physical inspection. The cost savings associated with the faults were estimated assuming conservative annualised unit costs of associated energy and surveyed air flow rates through each AHU.

subsequently be engaged in with the same pilot sites, and possible other sites depending on the stage of development. This will entail the sites monitoring the tool themselves to garner user feedback on the operation of the AFDD tool.

5.3 Results 5.3.1

Energy Savings identified by the AFDD tool The AHUs on several of the sites have been analysed in detail using the AFDD tool and a number of faults have been detected. Table 2 provides an overview of the major confirmed faults identified to date across the pilot AHU’s and their associated energy consumption savings which totalled over €104,000 per S ite

No. of AHUs in Pilot

1

2

2

4

3

9

4

3

Faults identified by the FDD tool

Passing heating coil, stuck actuator Damaged dampers, low temperature, passing cooling coil Damaged dampers

damper

Annual savings opportunities (approx) €66k

supply

€23k €3k

Passing heating coil, design issues

€12k

TOTAL

Verification method

Physical (airside) by the authors Physical (airside) by the authors Physical (airside) by the authors Physical (airside) by the authors

survey survey survey survey

€104k

Table 2: Savings identified 5.3.2 Individual Site Fault Samples 5.3.2.1 Site 1: Passing Heating Coil Actuation Valve A temperature rise was picked up by the AFDD tool across both the heating and cooling coils though the cooling coil was recorded to be 26% open. There is no mix temp sensor in the pilot AHU under analysis, so a virtual mix temp was utilized. Hence there was a chance that the virtual mix temp could have been calculated too low in error if the return damper was stuck in position or leaking. An airside temperature profile analysis was undertaken on the AHU by the research team.

Surveyed figures are shown on the graphic from the AFDD tool in the black boxes in red text (Figure 3). A temperature could not be obtained between the heating and cooling coils due to space limitations in the unit. The cooling coil was thus manually isolated and the temperature rise across the coils observed for a period of 15 minutes thereafter A 10 oC rise across the “closed” heating coil was recorded over this period. The mixing box damper was also observed to be in a state of fatigue. Based on the survey results and the data recorded by the AFDD tool, savings of approximately €48,000 per annum (thermal and electricity) are achievable by replacing the heating coil actuation valve and repairing the mixing box

Figure 3: AFDD GUI displaying its output & verification survey results for Site 1 fault

damper/actuator. A stuck return air damper was also identified during the alpha test period by the AFDD tool with annual cost savings of over €18,000 possible through its remediation. The AFDD tool also identified a lower than expected supply air temperature set point as well as a number of control logic improvements. 5.3.2.2 Site 2: Leaking Return Damper A temperature rise was picked up by the AFDD tool across the heating and cooling coils when the heating coil actuation valve was showing closed (0%) and the cooling coil was operational (27%). A Maintenance inspection by the research team confirmed the presence of a leaking return air damper due to damage to some of its louvers. A saving of approximately €4,900 per annum is projected based on the data leveraged from operation by the AFDD tool and the site survey flow rate and air side temperature profile information. 5.3.2.3 Site 3: Control strategy Issue The AFDD tool identified the heating coil as operating when the supply temperature was less than the maximum mix temperature possible with minimum fresh air. The mix temperature was measured by site survey to be at a lower temperature than the sensor recorded. The fresh air damper actuator appeared to be closed to its minimum position from the information garnered from the BMS data. This lead to a conclusion that the mix temperature sensor was either faulty or the outside air damper was either damaged or the actuator was stuck in position at a more open position allowing a greater quantity of cold outside air into the mixing box. 5.3.2.4 Site 4: Passing Heating Coil Actuation Valve The AFDD tool identified a 5 oC temperature rise across a closed heating coil. The research team undertook a temperature profile analysis of the AHU at site 4. The cooling and heating coils were forced closed manually by the team and measurements were taken before and after the coils. A 5 oC temperature rise was observed as the AFDD tool had measured thus verifying the presence of a passing heating coil in the AHU. Savings of over €4,000 are achievable by replacing the actuation valve on the heating coil. A passing heating coil actuation valve was also identified on another AHU onsite.

6. Conclusions and Future Work The results from the Alpha testing phase of this project are very encouraging. A data access and processing application has been designed and installed on each of these sites. This application accesses and uploads data from each of four test sites to a central database each morning automatically. It integrates

successfully with four different types of BMS system with very different data archiving methods. A business layer calculates virtual data points and detects faults in each of the 18 test AHU’s across 4 pilot sites on a daily basis automatically. A GUI has been developed which displays the faults detected a day behind real time on each of the alpha test units at AHU level by fault frequency. Consequently, over €104,000 per annum of energy savings have been identified and verified by site survey across the four alpha test sites. These faults would most likely either not have been identified using the traditional maintenance practices of the companies in question or they would not have been identified unless an energy audit or review of the system were undertaken. Invariably this would have led to considerable energy waste. A commercial case therefore exists for the further development of this software. A number of general observations from the alpha test phase should be taken into account in the next phase of development and testing. • Manual interaction with the alpha test AHU’s was identified as a major cause of energy inefficiency in the AHU’s under observation. Supply and zone set points were found to be manually adjusted to overcome what were actually ineffective control issues or physical faults in the system. • It was difficult to obtain design details for most of the pilot AHUs. It is hence not practical to base any future analysis on design data as it is difficult to obtain. Based on the work undertaken in the alpha testing phase of this project, it is clear that a number of areas require further research and development in order to ensure the widespread application of a robust AFDD tool namely; •

•

•

•

The AFDD tool must be a generic, cross-platform solution to ensure that data from any BEMS can be accessed efficiently. The error threshold analysis in this AFDD tool must be firmly based on robust statistical uncertainty approaches in order to minimise false positives and false negatives. This AFDD tool must be capable of supporting both the commissioning and operational phases on AHU operation. It is critical that this AFDD tool be capable of prioritising fault actions based on their relative severity Faults will be split into cost and comfort categories and a costing algorithm will be developed to quantify the financial implications of each fault.

•

•

The diagnostic capability of the AFDD tool must be improved. It is envisioned that this will be achieved through the possible use of a rule engine such as Drools [16] encompassing a forward/backward chaining algorithm or through the development of a stripped down “engine”. The use of case based reasoning or pattern matching techniques are also thought to be plausible. Finally, and possibly most importantly, it is critical that this AFDD tool is tested on a large number of actual AHU’s in operating HVAC systems by the end users in a variety of buildings and organisations, with validated energy saving results, in order to build a commercial case for this process. Acknowledgements

This research was funded by Enterprise Ireland. The authors would also like to thank the Innovation for Irelands Energy Efficiency (i2e2) Technology Centre.

[10]

[11]

[12]

[13]

[14]

[15]

References [1]

[2]

[3]

[4]

[5] [6]

[7]

[8]

[9]

L. Pérez-Lombard, J. Ortiz, and C. Pout, “A review on buildings energy consumption information,” Energy Build., vol. 40, no. 3, pp. 394–398, 2008. “HVAC SWG - Spin I, 2007.” [Online]. Available: http://www.seai.ie/Your_Business/Large_Energy_Use rs/Special_Initiatives/Special_Working_Groups/HVA C_SWG_07/. [Accessed: 13-Sep-2011]. ISO - International Organization for Standardization and ISO - International Organization for Standardization, “ISO - International Organization for Standardization.” [Online]. Available: http://www.iso.org/iso/iso_catalogue/catalogue_tc/cat alogue_detail.htm?csnumber=25052. [Accessed: 13Sep-2011]. M. A. Piette, S. K. Kinney, and P. Haves, “Analysis of an information monitoring and diagnostic system to improve building operations,” Energy Build., vol. 33, no. 8, pp. 783–791, Oct. 2001. International Energy Agency, “Computer Aided Evaluation of HVAC system Performance,” 34, 2002. International Energy Agency, “Cost Effective Commissioning of Existing and Low Energy Buildings,” 47, 2010. Daniel Choinière and Maria Corsi, “A BEMS assisted commissioning tool to improve the energy performance of HVAC systems,” in ICEBO, 2003. L. Gordon, “Modelling user acceptance of building management systems,” Autom. Constr., vol. 11, no. 6, pp. 695–705, Oct. 2002. W.-S. Jang, W. M. Healy, and M. J. Skibniewski, “Wireless sensor networks as part of a web-based

[16]

building environmental monitoring system,” Autom. Constr., vol. 17, no. 6, pp. 729–736, Aug. 2008. H. Wang, Y. Chen, C. W. H. Chan, and J. Qin, “An online fault diagnosis tool of VAV terminals for building management and control systems,” Plan. Future Cities-Sel. Pap. 2010 Ecaade Conf., vol. 22, pp. 203–211, Mar. 2012. F. Xiao and S. Wang, “Progress and methodologies of lifecycle commissioning of HVAC systems to enhance building sustainability,” Renew. Sustain. Energy Rev., vol. 13, no. 5, pp. 1144–1149, Jun. 2009. International Energy Agency, “Commissioning of Building HVAC Systems for Improved Energy Performance,” 40, 2006. J. M. House, H. Vaézi-Néjad, and J. M. Whitcomb, “An Expert Rule Set for Fault Detection in Airhandling Units,” Ashrae Trans., vol. 107, no. 1, Jan. 2001. J. Schein, S. T. Bushby, N. S. Castro, and J. M. House, “A rule-based fault detection method for air handling units,” Energy Build., vol. 38, no. 12, pp. 1485–1492, Dec. 2006. H. Yang, S. Cho, C.-S. Tae, and M. Zaheeruddin, “Sequential rule based algorithms for temperature sensor fault detection in air handling units,” Energy Convers. Manag., vol. 49, no. 8, pp. 2291–2306, Aug. 2008. “Drools.” [Online]. Available: http://www.jboss.org/drools/.