A. Introduction

The Health and Safety Executive (HSE) are a government funded organisation. They help to ensure that:

• employers look after the health and safety of their employees;
  
• employees and the self-employed look after their own health and safety; and
  
• everyone takes care of the health and safety of others, for example, members of the public who may be affected by work activity.

Your task is to design a series of interactive web pages that will provide readers with relevant information about accident frequencies in the United Kingdom.

B. Your Task: Alternative Visualisations for HSE Accident Statistics

As part of their activities, the HSE publish a range of accident statistics. These can be accessed on-line from:

http://www.open.gov.uk/hse/hsestats.htm

To minimise network delays, a copy of this information has been cached locally on:

http://www.dcs.gla.ac.uk/~johnson/hse_mirror

The HSE's accident statistics provide a broad range of information about accident frequencies. For example, they provide information about accident rates in particular industries. Within an industry, they record the frequencies of particular types of accidents (eg, burns or crush injuries). The importance of this data cannot be underemphasised. It is intended to alert employers and regulators to the dangers that threaten everybody every day of their lives.

Hint: in order to read these documents you will need to use Adobe Acrobat. It should already be installed on your computer and should launch automatically when you use Netscape. If you have problems then click on the link below to obtain a free copy:

http://www.open.gov.uk/howto/acroread.htm

The following table provides an example of the information provided by the HSE. This excerpt comes from the fact sheet on injuries to employees within the office-based industry reported to local authorities from 1991-1997:

Kind of Accident	1991/2	1992/3	1993/4	1994/5	1995/6	1996/7	Total
Contact with moving machinery or material being maintained	6	4	3	4	4	8	29
Struck by moving inc. flying/falling object	13	9	23	15	11	60	131
Struck by moving vehicle	4	4	6	6	2	9	31
Strike against something fixed or stationary	4	3	6	6	11	27	57
Injured whilst handling, lifting or carrying	8	7	15	10	9	29	78
Slip, trip or fall on same level	137	124	121	123	123	184	812
Fall from a height	71	57	51	52	40	81	352
Drowning or asphyxiation	_	_	1	_	_	2	3
Exposure to or contact with a harmful substance	11	10	5	6	3	4	39
Exposure to an explosion	_	1	_	1	_	3	5
Contact with electricity or an electrical charge	1	3	3	2	2	6	17
Injured by an animal	_	1	1	_	_	2	4
Acts of violence	_	_	_	_	_	15	15
Other kinds of accidents	3	2	2	6	_	3	16
Total	260	225	237	231	205	433	1,591

This format is appropriate because it provides readers with the specific data for categories of accidents within the UK. There are a number of problems, however. Firstly, it can be difficult to gain an overview of the more detailed information that is presented in the statistical reports. Secondly, it can be difficult to identify trend information where several different types of accidents may all be increasing, or decreasing, in frequency. Similarly, it may be difficult to determine whether particular types of accidents are increasing or decreasing against a general trend in the total number of accidents across all categories. Thirdly, these tables may be difficult for some members of the general public to understand. This is important because a deeper understanding of the relative frequency of particular accidents might help people to alter their behaviour appropriately. There are a range of graphs and charts for the HSE statistics on:

http://www.open.gov.uk/hse/hsestats/ssb9798.pdf.

These already solve many of the visualisation problems mentioned above but they could be made more interactive through the use of Java and AWT. For example, animations might be used to show how accident frequencies have altered year by year. Hypertext links might be used to connect statistical information with the textual commentaries that accompany the statistics. Your task is to design a set of web pages that will provide a particular group of users with graphical overviews and summary information about some of the accident data that is provided by the HSE. These pages can be developed in any way that you think supports the organisation's mission statement in section A.

C. The Design Brief

1. who are the potential users of the new pages?
   The HSE accident statistics are important form many people in many different industries. In particular, they may be cited by developers and designers to show that a proposed new system provides adequate protection against common risks. They are cited by lawyers during litigation. They are used by the media. They are also of interest to the general public. You must decide which one of these groups your system will support. (Hint: it may be easiest to select "the general public". If you do this, however, you will have to make assumptions about their expertise and previous exposure to human computer interfaces in general and graphical presentation techniques in particular).
   
   
2. what tasks might they want to perform with the new pages?
   Each one of the groups mentioned above will have particular tasks that they want to accomplish using the new systems. For example, designers may be interested about the risks associated with particular activities in particular industries. Other users may have less well specified requirements and may want to gain an overview by browsing the information. You must clearly identify the tasks of the user group that you are concerned with AND you must tailor your design to support those tasks.
   
   
3. what subset of the HSE data will you support?
   Given the constraints on this project, you will not have time to support all of the data provided by the HSE. You must, therefore, identify a subset of the data that will be sufficient to demonstrate your presentation techniques and provide sufficient data to verify that your software will work as intended.

The second stage of the project involves the development of initial designs and presentation formats. These should be as innovative as possible whilst still supporting the tasks that were identified in the previous stage of the project. It is possible to think of many different 2D and 3D visualisations. These include simple bar charts or graphs to show trend information. They also include more creative solutions. A barchart might be superimposed on a map to illustrate the number of accidents occuring in particular regions. A picture of a human body might be used to illustrate the number of injuries to particular limbs. One recent suggestion was that those limbs might be scaled to reflect the number of injuries to that part of the body; the head might appear twice as large as an arm if that part of the body sustained twice as many injuries. Clearly, the development of a suitable prototype requires design skills, creativity and a careful regard for the users' tasks. You may choose to develop a number of pencil and paper prototypes to test out your ideas. These should be subjected to some formative evaluation (see below). To help you, Ben Shneiderman's research group at the University of Maryland maintains a website that contains a vast range of potential visualisations on:

http://www.otal.umd.edu/Olive/

The third stage involves implementation. It is VERY important that your web pages are easy to maintain. In particular, it should be possible to update the visualisations with minimal effort as each year's statistics are compiled. (Hint: see Judy Bishop's use of the PARAM facility of applets in Java Gently, Addison Wesley, 2nd Edition, 1998, pages 348-349. Alternatively, you may choose to read in the data from a specially formatted file). Your solution must make use of AWT (see section A).

Finally, summative and formative evaluation must be conducted to test out whether your ideas. You must use a recognised technique described in the lecture notes and you must document your results in the final report.

D. Assessment Criteria and Submission Details

This exercise is degree assessed. It contributes 30% to the total marks associated with this course. The body of the report should not exceed twenty A4 pages. The report must be printed out and must be submitted in a secure binder (i.e., one that will keep the pages together and in the correct order). It must include:

1. A title page containing the name of the student as well as their contact details (email address etc);
   
2. A table of contents and appropriate page numbers;
   
3. A section on the design of the system. This should include some consideration of alternative approaches and a considered justification of the reasons why you build the system in the way that you did. (You might use the Questions Options and Criteria - QOC diagrams that have been mentioned in the lectures);
   
4. An implementation section. This should describe the techniques that you used during the development of your system. You can assume that the reader is familiar with AWT and Java. Please only mention areas of the code that are particularly complex or that require further work;
   
5. An evaluation section. This must consider both formative and summative testing. Formative evaluation might include user testing of pencil and paper prototypes and of partial implementations. Summative evaluation describes the testing of the final system.

1. the listing of any code used during the implementation together with suitable acknowledgements for the source of code that has been borrowed from other programmers;
   
2. internal documentation (maximum two sides). These are notes that are intended to provide guidance for the programmers who must maintain or develop you system. They should briefly describe the classes that you have implemented or re-used from other sources. They should also take the reader through any particularly significant methods (eg, event handlers) that are used.
   
3. external documentation (maximum two sides). This is the written user manual. You may decide that such a manual is not necessary for a web-based system. In this case, the manual should be aimed at the person who will generate the html to maintain the pages. You should not assume that these people will have any knowledge of Java.

The following marking scheme will be applied:

• 15 for the design;
• 15 for the implementation;
• 10 for the evaluation;
• 5 for the user documentation (on-line and off-line help);
• 5 for the technical documentation.

You may use any suitable programming mechanism to satisfy the following requirements. Suitable mechanisms include VRML, Java3D and Swing. The only caveat is that you must make use of AWT (1.0) or (1.1) at some point in your implementation.

E. Summary

This exercise provides you with a realistic design task. You are designing a real product for a real client that has the potential to be of practical benefit beyond the duration of your course. The accident statistics provided by the HSE help to ensure the safety of the general public in the UK and beyond. The visualisation and presentation of those statistics in an appropriate format is a challenging problem. However, if you come up with a suitable solution these is the possibility that the HSE will use your programs. This, in turn, will provide tangible benefits to society. It may also support your future career development.

You have a wide range of possible options and your task is to choose the best one that you can come up with. Please remember, however, that you have limited time and resources to complete your work. The marks will be based on the written submission. However, everyone will also be expected to present their solution during the final practical session of this course (Tuesday 16th March). An international panel of HCI experts will be asked to judge the best design solutions and prizes will be awarded to the winner and runners up.

Back to the main index...