Future Developments in Geographical Agent‐Based Models: Challenges and Opportunities

Abstract

Despite reaching a point of acceptance as a research tool across the geographical and social sciences, there remain significant methodological challenges for agent‐based models. These include recognizing and simulating emergent phenomena, agent representation, construction of behavioral rules, and calibration and validation. While advances in individual‐level data and computing power have opened up new research avenues, they have also brought with them a new set of challenges. This article reviews some of the challenges that the field has faced, the opportunities available to advance the state‐of‐the‐art, and the outlook for the field over the next decade. We argue that although agent‐based models continue to have enormous promise as a means of developing dynamic spatial simulations, the field needs to fully embrace the potential offered by approaches from machine learning to allow us to fully broaden and deepen our understanding of geographical systems.

Big-data approaches lead to an increased understanding of the ecology of animal movement

“…One promising approach, recently proposed for related challenges in geographical, social, and computer sciences, combines computationally demanding agent-based models and data demanding deep learning methods to decode hidden mechanisms from high-throughput data (89,90). Agent-based models can reveal the emergence of system-level patterns from local-level behaviors and interactions of system components (91).…”

“…Using genetic algorithms, initial candidate rulesets for individual decisionmaking can evolve into a robust ruleset that is able to reproduce the unique range and quality of spatial and temporal patterns in high-throughput data ("reinforcement learning" (89)]. Such patterns can be revealed by applying machine learning methods, including neural networks and deep learning (90). The combination of multiple patterns in highthroughput datasets at different hierarchical levels and scales leads to unprecedented model robustness, optimized model complexity, and reduced uncertainty (91).…”

Agent-based modelling for Urban Analytics: State of the art and challenges

“…Fortunately, a body of work has developed in fields that require the integration of real-time data to make more accurate short-term predictions, such as in meteorology [62], under the banner of 'data assimilation'. Broadly, data assimilation is starting to be recognised as an important barrier and opportunity for ABMs [41,43], and some effort in the group has been devoted to solving data assimilation challenges for ABM. Ultimately, if successful, data assimilation will allow ABMs to make better use of many of the 'novel' data sets that are becoming available, adapting in real-time as the social systems that they are tasked with simulating evolve (see Figure 6).…”

Incorporating E‐commerce into Retail Location Models

“…Third the well‐understood fundamentals of the SIM are extremely well fit for a first attempt to model e‐commerce dynamics in the retail sector. Such attempts require a clear validation of assumptions and parameters (Murray 2021), something which becomes less straightforward in more complex methods (Heppenstall et al 2021). Finally, the SIM has a proven track record in retail analytics and, as highlighted above, recent academic contributions (e.g., Siła‐Nowicka and Fotheringham 2019; Waddington et al 2019) are still developing novel innovations in this type of model.…”