Home > Research > Researchers > Fred Ingram

Fred Ingram

Research student

Fred Ingram

Lancaster University

The Management School




Most of my professional life involved Informational Technology, with over thirty years’ experience in Business Analysis, Project Management, Systems Analysis & Design, Data Warehousing, Programming, etc.  However, in the more recent past I spent 14 years teaching Mathematics and ICT at a UK secondary school.  I recently submitted my thesis, having being a full-time, fully-funded, doctoral student.  I now intend working on agent-based solutions to problems in any sector of industry, business etc. (although AI applications are of particular interest).

Thesis Outline

The primary goal of this research was to determine how and to what extent an agent-based model and simulation could adequately represent the behaviours of, and interactions between, students, a teacher and teaching assistant in classroom lessons at a UK secondary school. 

Following an in-depth case study at one school, an agent-based model of classroom lessons was formulated.  A lesson event recording tool was developed and used to record a variety of student and teacher activities.  These data were used to calibrate and validate the simulation model.  Several metrics were identified or constructed to characterise lessons and participants’ behaviour in order that the results of simulations could be compared and evaluated against each other and against the empirical data.  The data provided relative frequencies and durations for the various activity states of the students, teacher and teaching assistant.  The frequencies were used as probabilities to drive agent behaviour, with agent interactions controlled by sets of procedural rules, one set for each agent type.

The simulation model was found to generate plausible sequences of agent-level interactions for whole lessons, with overall lesson metrics that compared satisfactorily with the empirical lessons.  The model rules provided causal chains of explanations for outcomes, leading to deeper understanding of the mechanics of the complex classroom lesson system.

The conclusion was that an agent-based simulation model could represent classroom dynamics to the extent that it could be a useful lesson analytics tool, enabling teachers and educators to explore the mechanics and consequences of lesson interventions, including how they might affect overall class productivity and disruption.


BSc (Hons) Computer Science, MSc Information Systems, Cert. Ed. Post Compulsory, MA Education, BSc (Hons) Mathematics

  • Published

    Simulating classroom lessons: an agent-based attempt

    Ingram, F. & Brooks, R. J., 21/03/2018, Proceedings of the SW18: The OR Society Simulation Workshop. Anagnostou, A., Fakhimi, M., Meskarian, R. & Robertson, D. (eds.). Operational Research Society, p. 230-240 11 p.

    Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

View all (1) »

View all (5) »