This presentation including joint work with Alberto Franco, was presented at the IFORS (International Federation of Operational Research Societies) conference in Quebec City, QC, Canada. We present two different agent-based models for simulating human behavior.
We use the example of group decision making.
The first model uses a cognitive fitness landscape to model the quality of a decision, where participants compare their decision with their nearest neighbor. The decision is based on an external comparison.
The second model uses an internal comparison of a decision with the next best alternative. The model is based on the psychological concept of hidden profiles, where participants only make the best decision by sharing information with the group.
We will introduce a single capital floor, set at £100,000, more than four times the current means test threshold. This will ensure that, no matter how large the cost of care turns out to be, people will always retain at least £100,000 of their savings and assets, including value in the family home.
A quick calculation on the effective ‘Tax’ (Care Fees as a Percentage of Initial Wealth) shows the following distribution of Tax Rates. On the x axis is initial wealth (the value of your house plus any savings), and on the y axis is the Tax Rate. The Conservative Party have since augmented this plan with a proposed cap (consultation to come).
Values used for Care Fees: £20,000, £40,000, £60,000, £80,000; £100,000. Values used for Initial Wealth: £0, £100,000, …, £1,000,000. The trend continues downwards after this figure.
As many have pointed out, this affects individuals with initial wealth just over £100,000 proportionately far more than those with higher initial wealth. More detail on policy options for funding social care can be found in the Dilnot Commission report and a summary of their proposals is shown below:
The ‘℮’ symbol, or the ‘e-mark’ is a symbol you will see on packaging such as tins or packets in Europe. Millions of us will see this symbol every day, but what does it actually mean?
The raison d’etre for the e-mark comes from the problem of selling goods to the public. We would all like to think that we are getting what we pay for, but does that mean we should always get what we pay for?
Well, if you use the ℮-mark, then no. And yes if you don’t. So you use the ℮-mark. By doing so, some of us are short-changed, but, on average, we shouldn’t be.
The e-mark was introduced in 1976 by the legislation known by the snappy title of ‘Council Directive 76/211/EEC of 20 January 1976 on the approximation of the laws of the Member States relating to the making-up by weight or by volume of certain prepackaged products’.
This sets out a nominal value of a product. This means that, on average, we should not receive less than the value stated before the e-mark. But we would be really annoyed if we received, say, nothing, and someone else received twice the nominal amount. So, the concept of tolerable negative error was introduced at the same time, to set out the minimum legal amount that each packet or tin or container should contain. The idea is that only a few containers can weigh less than the declared value less the tolerable negative error (but none can be twice the tolerable negative error… that would be, well, intolerable).
In packets from 5 grams to 10 kilogrammes, the tolerable negative error varies from 9% (quite a lot) to 1.5% (not such a lot), the rationale being that it is easier to measure larger values with greater accuracy.
Eight Mile, epitomized by Eminem in the film of the same name, is a street in Detroit that marks the boundary between the majority white northern suburbs and the majority black neighborhoods closer to the inner city.
But what causes this segregation in the first place?
Hypothesis 1: The Central Planner
In Detroit’s case, as with many cities across the USA, it was, in part, due to the zoning of the city by the federal Home Owners’ Loan Corporation, which zoned the city into areas of risk, meaning that banks were indirectly encouraged to develop outer suburbs while not offering mortgages to inner city properties. This lead to wealthier, generally white, residents moving to the suburbs.
Indeed, physical barriers, such at the Detroit Wall, also known as the Eight Mile Wall, were built to separate majority black and majority white neighborhoods.
The legacy of these zones live on today, as seen in the map below from the 2010 US Census. The dividing line between the green (black) areas and the blue (white) areas is Eight Mile Road.
So, segregation exists, and is caused by a central actor. But is there an alternative explanation?
Alternative Hypothesis: Emergence
In 1971, Thomas Schelling set out to model the phenomenon, not by assuming a central planner, but by modelling the interactions of individuals.
Thomas Schelling’s model was this. Assume individuals are placed in a grid, similar to being located on a chess board. Allow individuals who are in a local minority to move. In the example below, the blue circle is in a minority (with 5 out of its 6 neighbors being a different color), and according to the rules of the model, is unhappy. It could decide to move to the vacant square to the north-west, but it would still be in a local minority (with 4 out of 6 neighbors being a different color) and would remain unhappy. So instead, it chooses the space to the south west where 3 out of its 6 neighbors are of the same color, and not being in a minority, it settles there.
Schelling, perhaps without knowing it, introduced agent-based modelling, where, instead of modelling the system as a whole, the modelling of individual agents enables us to see the emergence of macro-level properties, in this case segregation, via the modelling of micro-level (local) interactions.
We can see the effect of micro-level interactions causing macro-level segregation in the model below (developed by Duncan Robertson after Wilensky after Schelling). Each individual, or agent, decides whether they are unhappy or happy; if they are unhappy, they search until they find a vacant location where they will become happy. This continues until all individuals attain happiness.
So, perhaps segregation is not imposed, but is down to us. Or maybe, in reality, it’s a little bit of both.
Please do get in touch if you would like to discuss building or working with agent-based models.
In 2003, as part of a project to document coastal erosion in California, the following photo was posted on the californiacoastline.org website. It is a picture of a beach, some cliffs, lots of trees, and a rather nice house complete with swimming pool, that turns out to be the home of one Barbara Streisand.
For those not in the know, Barbara Streisand, is, in the words of her lawyer, a ‘renowned singer, actress, movie director, composer, and producer’.
Now, it turns out that Barbara Streisand values her privacy. To be specific, she puts a value of at least $10,000,000 on it. When the claim was filed, six people had downloaded the image. However, when Barbara Streisand issues a claim in the LA courts (in her own name) confirming that she lives in the nice house with the swimming pool, court reporters start twitching their notebooks. And so, it came to pass that once the lawsuit was publicised, everyone wants to know what Barbara Streisand’s house looks like. Nearly half a million in the first month.
The Streisand Effect, as it has been dubbed, is an example of unintended consequences. By planning to do one thing (suppress an image), you do the exact opposite.
The addendum to the story is that Barbara Streisand had a resurgence in popularity, culminating in the nearly 100-million downloaded song Barbara Streisand by the popular beat combo Duck Sauce.
Maybe this was, after all, a masterplan to take advantage of the unintended consequences of unintended consequences.
Agent-Based Modeling and Simulation is the first in the Operational Research Society’s OR Essentials series. OR Essentials brings together multidisicpinary research from the management, decision, and computer sciences. This edition within the series is edited by Simon Taylor, who is co-founder of the Journal of Simulation, also published by the OR Society.
The edited book is divided into 14 chapters, and the bulk of its contents convers the application of agent based modelling (ABM) to specific problem domains. The book adds to this by introducing agent-based modelling as a technique, and setting it in context with other simulation approaches. A very helpful chapter by Macal and North of Argonne National Laboratory offers a tutorial on what agent-based modelling is, focusing on the autonomy and interconnectedness of agents, and showing how agent-based models should be built. The book ends with thoughtful chapters on a testing framework for ABM (Gürcan, Dikenelli, Bernon), and a comparison with discrete-event simulation by Brailsford, elegantly closing the package opened by Taylor’s introduction comparing ABM with system dynamics and discrete event simulation within the context of modelling and simulation more generally.
The academic rigour of the book is confirmed by each article being reprinted from published articles from the Journal of Simulation. The book brings together several excellent examples of agent-based modelling, together with a very clear understanding of how ABM fits in with more traditional simulation techniques such as DES (Discrete Event Simulation) and SD (System Dynamics) – both Talyor and Brailsford show how and when ABM should be used. Macal and North offer a very useful tutorial for understanding the building blocks of an ABM simulation, while Heath and Hill show ABM’s evolution from cellular automata and complexity science through complex adaptive systems.
Domain specific chapters cover applications in the management of hospital-acquired infection (Meng, Davies, Hardy, and Hawkey); product diffusion of a novel biomass fuel (Günther, Stummer, Wakolbinger, Wildpaner); urban evacuation (Chen, Zhan); people management (Siebers, Aickelin, Celia, Clegg); pharmaceutical supply chains (Jetly, Rossetti, Handfield); workflow scheduling (Merdan, Moser, Sunindyo, Biffl, Vrba); credit risk (Jonsson); and historical infantry tactics (Rubio-Campillo, Cela, Cardona).
Agent-Based Modeling and Simulation offers a very useful collection of applications of ABM, and showcases how ABM can be successfully incorporated in to mainstream, published research. The contributions to the book are diverse, and from internationally regarded scholars. It is also useful to see the diverse ways that agent-based modelling research is presented, from some papers that show code, some that show running models, and some that do not show the model or code but instead describe results.
The glue that binds the book is methodological. Seeing how ABM has been used in diverse application areas is important, given the trans-disciplinary nature of the approach. It is an excellent introduction into agent-based modelling within a wide range of business and operations applications, and should be read by scholars and practitioners alike.
For new scholars, or for practitioners interested in learning about strategic management, it can be difficult to know where to start. The following is a list of the key journals in strategy together with some thoughts on their particular strengths.
Strategic Management Journal – perhaps the premier journal within strategic management. This mainly uses statistical modelling to test hypotheses, although is open to other approaches.
Management Science – a very good journal, but one that covers many areas of management, and therefore can only devote a small number of articles to strategy papers.
Strategy Science – a new journal that combines the best of the above journals – and provides analytical approaches to strategy problems.
Agent-Based Models of a Banking Network as an Example of a Turbulent Environment: The Deliberate vs Emergent Strategy Debate Revisited is a paper that I wrote that uses an agent-based model to simulate the strategic positioning decisions of firms within a competitive environment. It links firm strategy (particularly banking strategy) to the competitive environment: where should firms locate to gain most customers given that all firms are trying to achieve the same objective, that of maximizing market share / revenue / profits? It is an interesting question, and one that is very suited to modelling the dynamics of an industry using agent-based models.