Project Name

Operational excellence in IT service delivery

Tab navigation

The last two decades have seen unprecedented innovation in technology and its use in all functions of an enterprise. The global nature of business brought in additional challenges – the need for reliable, distributed, worldwide IT services – leading to large scale IT outsourcing. Consider the scale and scope of IBM’s IT outsourcing services – 1,000x clients, 1,000,000x IT assets and configurations in 100x data centers around the globe, 100,000x service personnel supporting these systems from global delivery centers, 1,000,000x services requests requiring a variety of competencies and skill levels, etc. This business is our lab and our research is to drive operational excellence – quality, productivity and client value in this complex and highly dynamic environment.

The research spans all aspects of service delivery:

  • Standardized processes, tools and optimal staffing to provide seamless service from delivery centers across the world. Address the challenge of identifying and modeling the relationships between IT function, service levels, human agents and business value generated.
  • Sophisticated infrastructure management to ensure compliance and optimal utilization. Comprehensive configuration management that drives key compliance (e.g. security, patch management) and health check processes.
  • Quality control and risk management for superior contract performance. Proactive defect prevention, continual improvement, predictive analytics to identify financial performance and identify/quantify key risks.
  • Big data and advanced analytics to glean insights and drive benefits to both IBM and our clients. Including In-depth insights to client’s IT landscape, performance, utilization, compliance, etc. Insights to recommend opportunities for IT optimization, transformation.

Highlighted projects:

1. Analytics Management System

Develop models for infrastructure impact on delivery quality and productivity and deploy to drive continuous improvement. The project aims to improve deployment scalability of the services delivery analytics by combining precise descriptive insights with predictive/prescriptive analytics over various globally available data sources, such as Configuration information available in Global Asset and Configuration Data Warehouse (GACDW), Ticket and Dispatch work orders available in Service Management Information Warehouse (SMIW) etc.

Analytics Management System

2. Lights Out - False Ticket Elimination

Develop and deploy analytics to reduce false alerts and apply optimal alert thresholding. The project analyzes the incident tickets to identify "non-actionable" alerts which lead to these tickets and automatically produces recommendations for tuning the alert thresholds to eliminate these without affecting the actionable alerts.

Lights Out - False Ticket Elimination

3. Financial Risk Analytics

Leverage the contract performance information captured in various questionnaires to predict contract risk estimation, including financial performance prediction and risk mitigation.

Financial Risk Analytics

Highlighted publications:


  • An Integrated Framework for Optimizing Automatic Monitoring Systems in Large IT Infrastructures
    Liang Tang, Tao Li, Larisa Shwartz, Florian Pinel, Genady Grabarnik
    KDD 2013
  • Recommending Resolutions for Problems Identified by Monitoring
    Liang Tang, Tao Li, Larisa Shwartz, Genady Grabarnik
    IM 2013
  • Quality Improvement and Quantitative Modeling – Using Mashups for Human Error Prevention
    Carlos Raniery Paula dos Santos, Lisandro Zambenedetti Granville, Larisa Shwartz, Nikos Anerousis, David Loewenstern
    IM 2013
  • Robustness of Comparison Sequential Test for the Piloting in Service Delivery
    Yefim Haim Michlin, Genady Grabarnik, Larisa Shwartz, Ofer Shaham
    BDIM 2013
  • Financial Risk Analytics
    Sinem Guven Kaya, Shu Tao, Sherif Goma
  • A Statistical Machine Learning Apporach for Ticket Mining in IT Service Delivery
    Ee Ee Jan, Jian Ni, Niyu Ge, Naga Ayachitula, Xiaolan Zhang
  • Business Insights via Machine Learning for IT Service Delivery
    Ea Ee Jan
    Frontiers of Service’2013


  • Discovering lag intervals for temporal dependencies
    Liang Tang, Tao Li, Larisa Shwartz
    KDD 2012: 633-641
  • A Learning Method for Improving Quality of Service Infrastructure Management in New Technical Support Groups
    David Loewenstern, Florian Pinel, Larisa Shwartz, Maira Gatti, Ricardo Herrmann
    ICSOC 2012: 599-606
  • Domain-Independent Data Validation and Content Assistance as a Service
    Maira Gatti, Ricardo Herrmann, David Loewenstern, Florian Pinel, Larisa Shwartz
    ICWS 2012: 407-414
  • Optimizing system monitoring configurations for non-actionable alerts
    Liang Tang, Tao Li, Florian Pinel, Larisa Shwartz, Genady Grabarnik
    NOMS 2012: 34-42
  • A learning feature engineering method for task assignment
    David Loewenstern, Florian Pinel, Larisa Shwartz, Maira Gatti, Ricardo Herrmann, Victor Cavalcante
    NOMS 2012: 961-967
  • Designing pilot for operational innovation in IT service delivery
    Genady Grabarnik, Yefim Haim Michlin, Larisa Shwartz
    NOMS 2012: 1343-1351
  • SUITS: How to make a global IT service provider sustainable?
    Parijat Dube, Genady Grabarnik, Larisa Shwartz
    NOMS 2012: 1352-1359 (Best Paper Award, BDIM)
  • A dynamic request dispatching system for IT service management
    David Loewenstern, Yixin Diao
    CNSM 2012: 271-275
  • Analysis of operational data to improve performance in service delivery systems
    Yixin Diao, Aliza Heching
    CNSM 2012: 302-308
  • Closed loop performance management for service delivery systems.
    Yixin Diao, Aliza Heching
    NOMS 2012: 61-69
  • Change Risk Expert: Leveraging advanced classification and risk management techniques for systematic change failure reduction
    Sinem Guven Kaya, Barbu Catalain Mihail, Husseman Dirk, Wiesmann Dorothea
    NOMS 2012:795-809


  • Performance management and quantitative modeling of IT service processes using mashup patterns
    Carlos Raniery Paula dos Santos, Lisandro Zambenedetti Granville, Winnie Cheng, David Loewenstern, Larisa Shwartz, Nikos Anerousis
    CNSM 2011: 1-9
  • Framework for Measurement and Prevention of Human Error in Service Delivery
    Larisa Shwartz
    JSM 2011
  • Service-Request Distributions in Business Processes and Their Control-Oriented Applications
    Genady Grabarnik, Yefim Haim Michlin, Larisa Shwartz
    JSM 2011
  • Defining Monitoring Parameters for Better Service Delivery Cost
    Larisa Shwartz, Genady Grabarnik
    INFORMS Annual Meeting 2011
  • Comparison Measurement of Service Processes
    Yefim Haim Michlin, Genady Grabarnik, Larisa Shwartz
    INFORMS Annual Meeting 2011
  • Adaptive memory load management in cloud data centers.
    Haishan Wu, Asser N. Tantawi, Yixin Diao, Wenjie Wang
    IBM Journal of Research and Development 55(6): 5 (2011)
  • A Report on CNSM 2010
    Yixin Diao, Hanan Lutfiyya, Noura Limam, Raouf Boutaba,
    J. Network Syst. Manage. 19(1): 137-142 (2011)
  • Staffing optimization in complex service delivery systems
    Yixin Diao, Aliza Heching,
    CNSM 2011: 1-9
  • The cost of service quality in IT Outsourcing
    Nikos Anerousis, Yixin Diao, Aliza Heching
    Integrated Network Management 2011: 773-784
  • Modeling a complex global service delivery system
    Yixin Diao, Aliza Heching, David M. Northcutt, George Stark
    Winter Simulation Conference 2011: 690-702
  • A Real-time Risk Assessment and Mitigation Engine based on Dynamic Context
    Sinem Guven Kaya, Barbu Catalain Mihail


Arjun Natarajan:
Larisa (Laura) Shwartz:
Amitkumar M. (Amit) Paradkar: