PolicySpace: agent-based modeling for public policy analysis

I've posted before about the "Humans of Simulated New York", a comprehensive agent-based model (ABM) of city life that is being led by Francis Tseng.

On a similar vein, my colleague from Ipea Bernardo Furtado has been developing the PolicySpace project, an agent-based modelling platform for public policy analysis. According to Furtado:

"PolicySpace is an agent-based model, including families, citizens, residences, businesses, markets, taxes, mobility, and municipalities, that allows “what-if” questions. It is an in silico laboratory, of extremely low relative cost. Yet, it is flexible, adaptable, that anticipates trajectories and, quantitatively, measures horizontal effects across sectors, places and times. The book reviews the literature, explains concepts, and describes the methodology. It details the model, its parameters, and the full process. It validates the proposal and illustrates with applications."

The platform allows for the ex-ante evaluation/simulation of public policy alternatives in a way that takes into account the emergent complexity of the interactions between portions of society and institutions, in space and time. PolicySpace was originally designed for the Brazilian case but it is easily adaptable to other contexts. The code is written in Python, it is open source and the full code is available on Github. The platform is also modular, so it can expanded in a flexible way to gradually incorporate different aspects considered to be relevant for a variety of policy realms. For example, Francis Tseng is further expanding the PolicySpace platform to incorporate public and private urban transportation at fine spatial scale.

Earlier this year, Bernardo published a book where he presents a literature review of ABM and where he introduces, validates and demonstrates applications of PolicySpace. The book PDF is freely available both in English and in Portuguese.

Give Bernardo a shout if you would like to collaborate on the project, use it in your own applications or just give him some feedback.  He is co-organizing a special issue on Complexity Science and Public Policy, so some of you might be interested in that as well.

image credit: Francis Tseng and Bernardo Furtado

A review of 85 Agent-Based Modelling platforms and tools

A recent paper has conducted a comprehensive literature survey comparing 85 Agent-Based Modelling platforms and tools according to the easy of development (simple-moderate-hard) as well as their capabilities (light-weight to extreme-scale). Via Danilo Freire (Twitter)

Abar, S. et al (2017). Agent Based Modelling and Simulation tools: A review of the state-of-art software. Computer Science Review. Volume 24, May 2017, Pages 13-33

Abstract:

The key intent of this work is to present a comprehensive comparative literature survey of the state-of-art in software agent-based computing technology and its incorporation within the modelling and simulation domain. The original contribution of this survey is two-fold: (1) Present a concise characterization of almost the entire spectrum of agent-based modelling and simulation tools, thereby highlighting the salient features, merits, and shortcomings of such multi-faceted application software; this article covers eighty five agent-based toolkits that may assist the system designers and developers with common tasks, such as constructing agent-based models and portraying the real-time simulation outputs in tabular/graphical formats and visual recordings. (2) Provide a usable reference that aids engineers, researchers, learners and academicians in readily selecting an appropriate agent-based modelling and simulation toolkit for designing and developing their system models and prototypes, cognizant of both their expertise and those requirements of their application domain. In a nutshell, a significant synthesis of Agent Based Modelling and Simulation (ABMS) resources has been performed in this review that stimulates further investigation into this topic.

By the way, one of the large-scale models that is relatively easy to use is the UrbanSim model, developed by Paul Waddell's team at Berkeley and which is freely available on GitHub. MATSim scales well with really large simulations and it's also open-source available on GitHub.

click on the image to enlarge it or go read the paper :) 

Humans of Simulated New York: comprehensive ABM model of city life

Here is a very interesting project that aims towards building agent based models to simulate cities and the impacts of policies on them. The project is available on GitHub. Kudos to the authors, who are all on Twitter btw Francis Tseng, Fei Liu and Bernardo Furtado.

Tseng, F., Liu, F., & Furtado, B. A. (2017). Humans of Simulated New York (HOSNY): an exploratory comprehensive model of city life. arXiv preprint arXiv:1703.05240.

Abstract:

The model presented in this paper experiments with a comprehensive simulant agent in order to provide an exploratory platform in which simulation modelers may try alternative scenarios and participation in policy decision-making. The framework is built in a computationally distributed online format in which users can join in and visually explore the results. Modeled activity involves daily routine errands, such as shopping, visiting the doctor or engaging in the labor market. Further, agents make everyday decisions based on individual behavioral attributes and minimal requirements, according to social and contagion networks. Fully developed firms and governments are also included in the model allowing for taxes collection, production decisions, bankruptcy and change in ownership. The contributions to the literature are multifold. They include (a) a comprehensive model with detailing of the agents and firms' activities and processes and original use of simultaneously (b) reinforcement learning for firm pricing and demand allocation; (c) social contagion for disease spreading and social network for hiring opportunities; and (d) Bayesian networks for demographic-like generation of agents. All of that within a (e) visually rich environment and multiple use of databases. Hence, the model provides a comprehensive framework from where interactions among citizens, firms and governments can be easily explored allowing for learning and visualization of policies and scenarios.

credit: Francis Tseng