A SIMULATION SYSTEM TO AID IN THE APPRECIATION OF INSURANCE RISK ASSESSMENT

P. J. Lawrence and T. G. McDonald

Faculty of Business and Law

Deakin University

Melbourne, Victoria

pjl@deakin.edu.au, tomxx@deakin.edu.au

Abstract

Deakin University conducts graduate and undergraduate programs in insurance. One of the main focuses of these courses is the underwriting of commercial insurance risks which involves the identification of the risks themselves. As many students have never been inside a factory they have difficulty appreciating the risks that need to be considered. The traditional approach to teaching risk relied heavily on the presentation of theoretical material which was then applied to written descriptions of risks. The alternative presented here is a Windows based program which allows the student to 'see' the risk being considered through photographs linked to a plan of the factory under consideration. The program requires the student to build an 'index' obtaining information from the plan, report and photos contained in the program. The index gives an indication of the degree of hazard presented by the factory. The program has proven to be very popular with and beneficial to the students.

1. Introduction

For over twenty years, in the field of Insurance, Deakin University and its antecedent institutions have been the principal source of tertiary education in Australia. Currently courses are offered at both graduate and undergraduate levels and one of the principal foci of these courses is the underwriting of commercial insurance risks.

2. The Problem

Traditional instruction in this area relies heavily upon the presentation of theoretical material either by way of lectures or by texts. The students are told of the principal features that must be considered in assessing risks under a range of headings and they apply this knowledge to risks presented in written case studies. The major problem that emerges from this situation has been indicated in surveys done of students in recent years which reveal that hardly any students have been inside a commercial risk. This makes it somewhat difficult for them to properly appreciate the assessment of risks.

Theories of the learning process indicate that the learner is actively involved in constructing and reconstructing their knowledge base (Anderson, 1983; Rumelhart & Norman, 1981; Rumelhart 1980). Traditionally the learner was viewed as an 'empty box' into which knowledge could be poured but these newer approaches stress the active role of the learner (Shariffudin, Kassim, Buang, 1994).

The ideal approach would be for the students to be taken to a plant and there allowed to undertake a survey for themselves attempting to apply the knowledge gained from the materials provided. However, there are a number of problems associated with this approach.

3. Requirements

To be in a position to meet all of these needs it would be necessary to have something that would satisfy the following requirements:

4. The Solution

The solution satisfying these requirements has been developed in the form of a simulation system which will be described in this paper.

Simulations are widely used in commercial education in Australia (McKenna 1991), the U.S. (Faria 1987) and Britain (Burgess, 1991). Commercial availability of such programs focus more on areas such as production management (Greenlaw, Hottenstien and Chu, 1991), manufacturing (Cotter and Fritzsche, 1991; Lawrence 1995) or strategic portfolio management (Priesmeyer, 1992). Although insurance simulations do exist (Lawrence and McDonald 1995) there does not appear to be a suitable program for the insurance risk assessment and for this reason the Property Insurance Risk Assessment program was developed.

The system is a Windows based interactive program that presents the student with written and pictorial descriptions of a factory. The program can be divided into three parts:

4.1 The Information

The student is presented with:

4.2 Risk Index

The students' task is to build a hazard index. The index is one that has evolved over a number of years. It started with Dr. Greber who developed an index for the Swiss Fire Protection Agency to assist Fire Brigades in assessing the fire of premises (Swiss Society of Engineers & Architects 1984). The index has been altered and expanded by many different people around the world and was again altered to suit the needs in this case. The index originally was for assessing the fire risk and has now been expanded to include a number of other perils that are commonly offered on insurance covers in Australia.

The index is quite detailed and because of the large amount of information needed to build it is not used by insurers in assessing risks. From the insurers viewpoint the cost of gathering a large amount of information is not offset by comparable benefits. The reduced amount of information they gather at present is adequate. The detailed index is ideal from a teaching viewpoint, however, as the student learns from considering each item on the index, the factors that distinguish risks and can see the factors that cause the index to rise. In other words those factors that make one risk more hazardous than another.

The index can also show how a risk can be improved by seeing those items that cause the index to rise.

4.3 The Assessment

When a student has finished building the index they can check on how they fared by going to the assessment section. In this section they are shown the individual components of the index and both their response and the correct response. Also shown is the location, in all of the materials provided, where the correct response can be found. In this way they are able to return to the program and look up the areas in which they did not provide the correct answer and see where they have missed the item.

5. The Program In Detail

This program was written in Authorware. Other programs using similar concepts, but in different contexts, have also been written in both Director and Microsoft Visual Basic.

5.1 The Plan

With the information section the principal need is to present the information in an easily accessible manner for the student. The report is text based and provides the student with a fairly good picture of the operation of the factory. The plan starts to give reality to the report. The photos, which are accessible from the plan, were taken from a range of angles. There are 30 photos in all. One set was taken from an elevated position looking down to the factory floor and was taken in a way such that when they are placed side by side they provide a view of three sides of the factory. A reduced version of this set of photos is placed on the same screen as the plan and serves to give the student a good overview of the factory and helps put into perspective the description given in the report. By clicking on any part of this panoramic picture the individual photograph making up that section is displayed in full and finer detail can be seen. Where greater detail is needed, clicking on that section of the photo provides a close up view.

A factor that had to be kept very much in mind when developing the program was the computer on which the program would ultimately be run. Student surveys had indicated that there were a number who were using 386SX machines. It also appeared that all were using SVGA display cards, which would be necessary to adequately display photos.

However, it became apparent when taking these surveys of students' computing equipment that a reasonable number did not have any idea of the type of computer they had and yet were quite prepared to answer a questionnaire without checking. This could perhaps have lead to false survey results which in turn can be misleading to anyone relying on those results.

The photos are a very important part of the program but they also present a problem as far as memory is concerned. The higher the resolution the easier it is to identify the detail which is a very important aspect of the program. On the other hand the higher the resolution the larger the program size which presents problems so far as the distribution, storage and running of the program is concerned. For example, using 24 bit colour depth the pictures could take up as much as 12Mb of storage space. On older, 386 machines such large files would be very slow to load. Using 256 colours the graphic files could be reduced to 4Mb. The highest resolution that a screen display will go to is about 75dpi. If this is reduced to 65dpi the picture quality is still reasonable and the memory size is further reduced to 2Mb. The time taken to load and run these images will reduce also.

Distribution is also a problem with the higher resolutions. The high colour resolution version of the program (12mb plus) would require about 11 floppy disks. By using compression programs this could be reduced to 4 - 5 disks. The 2mb file version, however, could be reduced to two floppy disks by using a compression program.

The use of CD-ROM would overcome the distribution and storage problems. In one respect the current price drops in CD-ROM read/write drives is very encouraging but the problem still remains that corporate buyers have not been significant buyers of CD-ROMs. While ABS figures show that 41% of home computers have CD-ROM drives anecdotal evidence tends to indicate that corporate buyers are moving more slowly in buying them. Many of these students are part time and are using company computers to run the program. It would seem that the use of floppy disks will need to continue at least until the availability of CD-ROMs is more widespread. With current versions of programs requiring increasingly large file sizes it would seem the number of disks on which programs need to be distributed will be on the increase.

The provision of close up photographs of different areas of the factory can be achieved to some extent by expanding the initial photo but this can only be done to a limited degree because the photos becomes grainy with quite modest expansions. The better approach is to take a fresh photograph of the close up area even though this means increasing the file size. The end result is much better.

5.2 The Index

The program builds three indexes. One for fire, one for explosion and one for extraneous perils. These three areas form the basis of cover provided in relation to most property insurance contracts. The reason for three separate indexes is that these exposures are quite separate and are rated separately by insurers. To merge them into one could produce a misleading result.

The index has been devised to be capable of application to all types of property. It therefore includes many different categories and a number of these may have no application on a specific property. Each category needs to be considered in every case and a decision made as to its applicability.

Within each category there are a number of choices and in some cases more than one can apply. The student needs to consider the choices in a given category then select the appropriate item(s). In considering how this section was to be presented to the student it was necessary to take into account that the student would need to look at a given section on the index then go to the program, look for features relating to that section, return to the index building section and finally enter the result. This would be a very cumbersome approach and one that would frustrate the student causing loss of interest in the exercise. It was for this reason that a hard copy was produced of the index and this was included in the student manual. With the hard copy the student is able to go to the plan and report, look for the categories mentioned on the hard copy and enter them. A further advantage with this approach is that if, while looking at a photo in relation to one category, the student notices a point that relates to another category they are able to enter that other matter on their hard copy also. In view of the large number of items to be considered this approach has proven to be the best.

When the hard copy sheets are completed the student can go to the index page, where the same categories and options also appear, and select the option decided upon. By this stage the student has selected their options so the best approach was thought to be one that allows the student to enter the data quickly. This was achieved by having each possible selection shown being a 'hotspot' and when it is clicked on that amount is added to the index. In this way the student does not have to type in any information but merely clicks on the desired item. This approach is beneficial to all but particularly to those who have limited keyboard skills. If an error is made in entering data in a section there is provision for erasing that section and redoing it. Each section has only 5 or 6 items in it so redoing a section does not involve a great amount of work.

The index builds as each item is entered. As the items can be changed easily the student can see, as an item is entered, the effect that item has on the overall index. The effect that the items have on the overall index also indicates to the student areas in which the exposures presented by the factory can be improved. Some features obviously cannot be changed. For example if the premises are of wood construction this will cause the index to rise sharply but it is not an area in which an alteration can be made. Alternatively, if a flammable varnish were being used on a manufactured product the index would rise but if a non flammable varnish were used the index would not rise. By reviewing the effect that items have on an index it is possible to see the ways in which the exposures a particular factory presents can be improved.

5.3 The Assessment

The assessment section is, in one respect, a duplication of the index. The assessment section has the same layout as the index section. By the time the student has completed the index and is ready for assessment they will have become fairly familiar with the layout of the index pages. By making the assessment layout the same it will be much easier for the student to follow. The assessment pages show the options that appeared on the index building pages as well as the student's response and the correct response. This way the student can see immediately those areas where their response differs. In addition, alongside each item the student is told where in the program to find the correct answer. In this way the student can go back and look at the area to see how the information was presented.

6. Student Responses

This program was produced and made available to the students early in the semester, in fact, an assignment was built around the program. The students were required to complete the program then write a paper on how the factory could be made a better risk.

The first response received from a student was an unsolicited phone call to say what a great concept they though this was. This attitude transpired to be quite widespread through all students. At the end of the semester they were sent a questionnaire asking for feedback and responses were very positive. A representative sample of specific student comments are set out below.

These are only a few of the comment made by students on the questionnaires and some of them could indicate that the attraction was more novelty value with the use of a very different medium. However, a stronger message is coming through and that is that being able to see the risk gives a much better appreciation and understanding of the risk and the topic of risk assessment in general.

There is, of course, no intention to suggest here that the program is perfect. In fact, because of the early positive reaction to the program, the students were invited on the questionnaire to identify any perceived shortcomings and to be as critical as possible. This was undertaken to improve future versions of the program. This process brought forward a number of constructive suggestions relating mainly to the ability to use short cuts to access different parts of the program. It was nonetheless encouraging to find that not one of the responses was in any way negative.

In addition, assessing student submissions for this assignments revealed a significant increase in their quality as compared to the paper based version of the exercise used in previous semesters. Student understanding of the topic and appreciation of the exposures seemed far better than in the past. While there is no doubt that this could be due to a range of factors, it is felt that this program has played a major role in that improved understanding and it is consequently intended to continue the use of the program including incorporation of the student suggested improvements.

7. References

Burgess, T. F. (1991), The Use of Computerised Management and Business Simulation in the United Kingdom, Simulation and Games, v22, n2, pp174-95.

Cotter, R. V. and Fritzsche, D. J., (1991), The Business Policy Game, 3rd Ed., Prentice Hall, Englewood Cliffs, USA.

European Conference of Fire Insurers (1980), Hazard Evaluation of Materials and Goods, European Conference of Fire Insurers Standing Technical Committee of the Fire Working Group, Paris, France.

Faria, A. J. (1987), A Survey of the Use of Business Games in Academia and in Business, Simulation and Games, v18, pp207-24.

Greenlaw, P.S., Hottenstein, M.P. and Chu, Chao-Hsien (1991), PROSIM, A Production Management Simulation, 2nd Ed., Harper Collins, USA.

Lawrence, P.J. (1995), The Didactic Industrial Management Simulation, Proceedings of ASCILITE 1995, Melbourne, Australia, Dec. 1995.

Lawrence, P.J. and McDonald, T. G. (1995), The General Insurance Market Instructional Computer Simulation, Proceedings of ASCILITE 1995, Melbourne, Australia, Dec. 1995

McKenna, R. J. (1991), Business Computerised Simulation: the Australian Experience, Simulation and Games, v22, n1, pp36-62.

Priesmeyer, H. R. (1992), Strategy! A Business Unit Simulation, 2nd Ed., South-Western Publishing Co., Cincinnati, USA.

Rumelhart & Norman, (1981). Analogical processes in learning. In J.R. Anderson (Ed.) Cognitive skills and their acquisition (pp 335 - 361). Hillsdale, N.J.: Lawerence Erlbaum.

Rumelhart, (1980). Schemata: The building blocks of cognition. In R. J. Spiro, B. C. Bruce & W. F. Brewer (Eds). Theoretical issues in reading comprehension (pp 33 - 59). Hillsdale, N.J: Lawerence Erlbaum.

Shariffudin, R.S.; Kassim. A.H; Buang, K. (1994), presented at the APITITE Conference, Brisbane, 1994.

Swiss Society of Engineers & Architects (1984), SIA Method of Hazard Evaluation.