Avishai Mandelbaum

Avishai Mandelbaum (born 1955) is an Israeli professor of operations research, statistics, and service engineering at the Technion — Israel Institute of Technology. He is best known for creating the Erlang-A model (M/M/N+M), which extends the classical Erlang C formula by incorporating customer patience and abandonment — a contribution that fundamentally changed how the contact center industry models staffing under real-world conditions. Mandelbaum popularized the term "Service Engineering" as an academic discipline and co-authored the landmark 2003 survey paper that defined call center operations research as a field.

Overview

Avishai Mandelbaum occupies a unique position at the intersection of queueing theory, statistics, and practical workforce management. While Agner Krarup Erlang gave the industry its foundational traffic models in the early twentieth century, those models assumed infinite customer patience — callers who would wait forever. Mandelbaum's Erlang-A model corrected this unrealistic assumption by adding an exponential patience distribution, enabling analysts to predict abandonment rates, waiting times, and required staffing levels with far greater accuracy. Every modern WFM tool that accounts for abandonment owes a direct intellectual debt to his research.

Beyond the Erlang-A model, Mandelbaum has been instrumental in building the theoretical infrastructure that connects academic queueing science to operational contact center management. His work on heavy-traffic approximations, data-driven service systems, and the Palm/Erlang-A framework has influenced both the scholarly literature and commercial WFM software design.

Early Life and Education

Mandelbaum earned a B.Sc. in Mathematics and Computer Science and an M.A. in Statistics, both summa cum laude, from Tel Aviv University. He completed his Ph.D. in Operations Research at Cornell University, where he studied under the influence of a strong applied probability tradition. After graduating in 1983, he joined the Graduate School of Business at Stanford University, where he remained on the faculty until 1991. He then returned to Israel to assume a professorship at the Technion, where he has been based ever since.

Career

Stanford University (1983–1991)

At Stanford, Mandelbaum established his early research program in stochastic processes and queueing theory. This period laid the groundwork for his later focus on service systems, as he developed mathematical tools for analyzing complex stochastic networks that would eventually be applied to call center operations.

Technion — Israel Institute of Technology (1991–Present)

Upon joining the Technion's Faculty of Industrial Engineering and Management (now the Faculty of Data and Decision Sciences), Mandelbaum built what became one of the world's leading research programs in service engineering. He holds the Benjamin and Florence Free Chair in Applied Probability and co-founded the Technion SEE Laboratory (Service Enterprise Engineering) in 2007, serving as its academic director since inception.

The SEE Lab became a hub for data-driven research on service systems, partnering with organizations including banks, hospitals, and government agencies to collect and analyze operational data at scale. This emphasis on empirical grounding — using real call-by-call data rather than purely theoretical models — distinguished Mandelbaum's approach and made his work directly applicable to WFM practice.

Chinese University of Hong Kong, Shenzhen

Mandelbaum has also held a visiting professorship at the Chinese University of Hong Kong, Shenzhen, at the School of Data Science, extending his research program into healthcare service engineering and data science applications.

Key Contributions

The Erlang-A Model

The Erlang-A model (M/M/N+M) is Mandelbaum's signature contribution. The classical Erlang C model assumes that all customers who arrive at a queue will wait indefinitely for service. In reality, customers abandon — they hang up, leave the queue, or switch channels. The Erlang-A model adds an exponential patience distribution to the standard M/M/N queue, creating a model that captures three simultaneous phenomena: arrivals, service, and abandonment.

This seemingly simple extension has profound practical implications. The Erlang-A model produces finite waiting times and finite queue lengths even when traffic intensity exceeds capacity — a situation where Erlang C predicts infinite queues. For WFM analysts, this means more realistic staffing recommendations, particularly during peak periods when abandonment rates are highest.

The Palm/Erlang-A Framework

Working with Opher Garnett and Martin Reiman, Mandelbaum developed the Palm/Erlang-A framework in a landmark 2002 paper published in Manufacturing & Service Operations Management. This framework provided closed-form approximations for key performance metrics (wait time, abandonment probability, service level) that are computationally tractable for real-world staffing calculations. The framework bridged the gap between theoretical queueing models and the practical needs of WFM software vendors who needed fast, accurate algorithms.

Service Engineering as a Discipline

Mandelbaum coined and championed the term "Service Engineering" to describe the systematic, data-driven design and management of service operations. His vision positioned call centers not merely as operational cost centers but as complex service systems amenable to rigorous engineering analysis. This framing elevated WFM from a scheduling task to an engineering discipline, influencing how universities structure their curricula and how practitioners conceptualize their work.

The 2003 Landmark Survey

Co-authored with Noah Gans and Ger Koole, "Telephone Call Centers: Tutorial, Review, and Research Prospects" (2003) became THE foundational reference for call center operations research. Published in Manufacturing & Service Operations Management, the 62-page survey mapped the entire landscape of call center OR — from forecasting and staffing to routing and performance measurement — and identified open research problems that shaped the field for the next two decades.

Heavy-Traffic Approximations

Mandelbaum's work on many-server heavy-traffic limits extended the Halfin-Whitt regime to systems with abandonment. His "Excursion-Based Universal Approximations for the Erlang-A Queue in Steady-State," co-authored with Itay Gurvich and Junfei Huang, won the Uri Rothblum Prize from the Operations Research Society of Israel (ORSIS) in 2014.

Legacy and Impact

Mandelbaum's influence on workforce management operates at multiple levels. At the theoretical level, the Erlang-A model corrected a fundamental limitation of Erlang C that had persisted for decades. At the practical level, his work provided the mathematical foundations that WFM software vendors incorporated into their staffing engines. At the disciplinary level, his championing of "Service Engineering" helped establish call center operations as a legitimate field of academic study.

The SEE Lab's emphasis on data-driven research — collecting and analyzing millions of call records from real service systems — set a standard for empirical rigor that influenced subsequent research programs worldwide. His students and collaborators have gone on to hold positions at leading universities and have continued to advance the field.

Awards and Recognition

• Uri Rothblum Prize for Outstanding Publication, ORSIS (2014)
• Technion Yanai Prize for Excellence in Academic Education
• Technion Taub Prize for Academic Excellence
• Fellow of the Institute for Operations Research and the Management Sciences (INFORMS)

Connection to Workforce Management

Every WFM system that models customer abandonment traces its intellectual lineage to Mandelbaum's Erlang-A model. When a workforce analyst uses an Erlang-A calculator to determine staffing requirements — and gets a different (usually lower) answer than Erlang C would provide — they are using Mandelbaum's mathematics. The waiting time distributions in modern WFM platforms, the patience parameters in simulation models, and the staffing algorithms that account for finite customer tolerance all derive from the theoretical framework he built.

His 2003 survey paper with Gans and Koole remains required reading for anyone seeking to understand the academic foundations of WFM, and his Service Engineering framework provides the conceptual vocabulary that connects queueing theory to operational practice.

Selected Publications

• Gans, N., Koole, G., and Mandelbaum, A. "Telephone Call Centers: Tutorial, Review, and Research Prospects." Manufacturing & Service Operations Management, 5(2), 2003, pp. 79–141.
• Garnett, O., Mandelbaum, A., and Reiman, M. "Designing a Call Center with Impatient Customers." Manufacturing & Service Operations Management, 4(3), 2002, pp. 208–227.
• Gurvich, I., Huang, J., and Mandelbaum, A. "Excursion-Based Universal Approximations for the Erlang-A Queue in Steady-State." Operations Research, 62(3), 2014, pp. 615–633.
• Mandelbaum, A. and Zeltyn, S. "Service Engineering in Action: The Palm/Erlang-A Queue, with Applications to Call Centers." In Advances in Services Innovations, Springer, 2007.
• Brown, L., Gans, N., Mandelbaum, A., et al. "Statistical Analysis of a Telephone Call Center: A Queueing-Science Perspective." Journal of the American Statistical Association, 100(469), 2005, pp. 36–50.

See Also

• Erlang-A
• Erlang C
• Abandonment Rate Modeling and Patience Distributions
• Waiting Time Distributions
• Queueing Theory Fundamentals
• Ger Koole
• Noah Gans
• Operations Research in Workforce Management