GEOG3150 - GIS, Geocomputation and Geoplanning - Semster 2

Semester 2 Project: Understanding Social Systems with Agent-Based Modelling

Aim of the Assessment

The aim of the project is for you to improve your understanding of the process of building an individual-level model and the value of modelling for understanding the present and predicting the future. It will also help to develop skills in designing, constructing and running models in NetLogo. It builds directly on the practical work that you have been doing (modelling a music festival).

Assignment weighting

This assignment contributes 30% to the final mark of this module

Assignment deadline

The deadline for submission of this assignment is 14:00 on Thursday 7th May 2015 (week 24)

The Scenario

The organisers of a local music festival are concerned with the amount of crime that takes place there. They have approached you to create a model of the festival and suggest ways in which they could intervene to prevent people from being mugged.

Your task is to experiment with an agent-based model that simulates the movements of visitors as they walk around the festival sites and muggers who will attempt to rob the visitors. The festival organisers have recruited a security company to reduce the amount of crime and you need to use your model to suggest the optimal locations for security booths. After running some experiments with your model you need to produce a report that explains your findings and provides policy suggestions for the festival organisers.

Assessment Details

This assessment is somewhat unusual in that there are two different options (A and B) that you can choose. One is predominantly a practical modelling task, the other involves writing a report (no extra programming).

Option A - Programming

Option A is mainly a practical modelling task. You will improve the festival model that you developed during the practical sessions to make it more realistic (a better representation how people really behave in a music festival). After making some improvements, you will then conduct experiments to find the optimal locations of the security booths. You are free to improve the model in any way that you wish, but some examples that you might like to implement include (in approximate order of difficulty):

  1. Create a graph to show the total number of muggings. At the moment the graph shows the rate.
  2. Make some visitors vegetarian, so that they will never travel to burger vans
  3. Make the visitor agents spend some time at their destination (different parts of the festival sites) when they reach it, rather than just arriving and then leaving immediately.
  4. Add a new button that makes all visitor agents go to the main stage and stay there for a number of iterations (this is to simulate the different type of behaviour exhibited by visitors when a band is playing).
  5. Change the behaviour of the muggers so that there is a probability that a mugging will be successful (at the moment all muggings will be successful, as long as the muggers are not near a security booth).
  6. Improve the behaviour of the muggers so that rather than walking around randomly, they go to the parts of the festival that have the most visitors.
  7. Use Links (or another mechanism) to allow friendship groups of visitors to move around together.

The model does not need to do all of those options; marks will be awarded progressively for the difficulty involved in adding each improvement. You are also encouraged to add your own improvements and can talk to Nick if you're not sure about how difficult or useful they will be.

Note: to get the very best marks you will need to show that you really understand how NetLogo works and are able to go beyond the work that we did during the practicals.


There are two deliverables for Option A:

  1. Your final NetLogo model. This will be assessed on:
    • its functionality - i.e. what it can do (this is the most important criteria);
    • the quality of the source code that you have written, i.e. the elegance of your code and the usefullness of the comments.
    Also, you should include the following information in the 'Info' tab:
    • a description of the changes that you have made with a justification;
    • thoughts about further improvements you would like to make given more time.
  2. A 500 word report outlining:
    • the experiments that explore the optimal locations of the security booths - based on your model where are the optimal locations?

Submission of assignment A.

A submission area will be created to allow both the report and NetLogo model to be submitted. This will appear in the 'Turnitin Submission Area' on the VLE.

Marking Criteria

DeliverableComponent% of GradeMarking Criteria
500 word report 20% Your report will be assessed according to the generic School of Geography marking criteria (a copy is available on the VLE (and also here for 2013/14). However, as the report is very short and more technical than a normal essay, a critical engagement with the literature is not required (although some references might be appropriate, particularly if you have adapted other peoples' source code).
NetLogo Program Model functionality 60% Your model will be assessed on the extent of the improvements that you have made. Models that are able to accomplish all of the suggested improvements above (or similar improvements that you have come up with yourself) will receive top marks.
NetLogo source code 20% Your source code will be assessed on the:
  • abundance and usefulness of comments
  • code layout and ease of understanding
  • code efficiency and elegance


Option B - Report

Option B requires no extra NetLogo development. The task is to use the model that you created during the practical sessions, run some experiments to explore the optimal locations for security booths, and write a 1,500 word report outlining your findings. As there is no extra programming this report needs to be much more comprehensive than the report for Option A.

The report can be structured in any way that you wish, but the following is an example of how you might choose to structure it and the content that might be included:

You should make reference to the ideas introduced in lectures (e.g. emergence, modelling behaviour, interaction, etc.) and show evidence of wider reading through the inclusion of relevant academic references. Also, remember that the report is for a lay reader who is unlikely to be familiar with agent-based modelling or other geocomputation techniques.


The only deliverable for Option B is the 1,500 word report.

Submission of assignment B.

The 1,500 word report should be submitted through Turnitin on the VLE as usual. See the 'Turnitin Submission Area' on the VLE.

Marking Criteria (Option B)

The generic School of Geography marking criteria will be used to mark the report. A copy is available on the VLE (and also here for 2013/14).

Other important info

Please note that for elements give a word limit, there is no + 10% leeway for word length. Penalties for going over length are outlined in the School's Code of Practice on Assessment (CoPA).

The VLE time stamp (which will be added to your submission when you upload the electronic version of your proposal) is definitive in terms of monitoring late submissions. You should ensure that you have acknowledgement from the system of submission of your work. If for any reason the system does not give you an acknowledgement that you have submitted your work, then you should report this immediately to the module conveyor by email and attach a copy of the assignment you were trying to upload.

Make sure you are aware of the deadlines, and please note that late submission will lead to the imposition of the normal University penalties; 5 marks are deducted for a submission that is made after the deadline (2 pm) on day 1 and any time until 2 pm on day 2. Thereafter students lose a further 5 marks for each subsequent period of 24 hours.

Assessment Return Date to students: Students will normally receive feedback, via the VLE, within 3 term-time weeks.

You are expected to use the University's version of the Harvard referencing style for your assignments for this module. Guidance on how to include citations within your text and how to reference different types of material using Harvard is provided here.

The above requirement is to ensure consistency across the University. However, for the taught postgraduate MSc degrees in geography, with regard to citations and referencing where there are more than two authors, please use the format (Author, et al., year) when citing a source within your main text. In the reference list, please write out all author surnames and their initials (rather than just the first author, et al. as in the guidance noted above). This is the format of Harvard which is used in journals relevant to our topic areas.

Skills@Library provides academic skills teaching and e-learning support so that students and researchers can effectively find, use, apply and present the information they need.

[School of Geography homepage] [Leeds University homepage]