Coursework Part A: @Risk Modelling
Coursework Part A: @Risk Modelling
this is a coursework for simulation module in msc risk management. i am attaching the guidelines on how to write this report and another document which tells what the course work is about.. this coursework required 1500 words and the working also on an excel sheet seperately which is mentioned in the documents also. it requires you to use @risk for this coursework.
MANG 6122 Simulation 2014-15
Coursework Part A: @Risk Modelling
Total marks for this part: 40%
Deadline for this part: Friday 12 December 2014
The coursework will be assessed according to the extent to which it indicates an understanding of relevant techniques and methods of use, and, where appropriate, the ability to apply them and the extent to which it displays independent thought and judgement. Marks will be given for clear explanations and diagrams and evidence of practical skills with spreadsheet modelling and with the use of @Risk.
1. You must upload on to the Blackboard site as “Part A – Report” a BRIEF report (1500 words maximum but less is fine) for the TSP board (see case study overleaf). Please label this file with your family name, e.g. mine would be “Brailsford.docx”. The report should be in the style of a management report (see website) and any technical terms should be clearly explained. You should present your business case if you decide to ask for extra resources, or argue why you feel it would not be worth spending any extra money. This report must be non-technical; you should explain in simple terms what your simulation shows, and why you are confident in your recommendations. DO NOT include any appendices or computer printout, although you can refer to the relevant worksheets in your Excel file. Any printout which is submitted will be disregarded.
2. You must hand a printed copy of this report in to Reception.
3. You must also upload an Excel file (an .XLSX file) on to the Blackboard site as “Part A – Model”. Please label this file with your family name, e.g. mine would be “Brailsford.xlsx”. The first worksheet in this Excel file must contain your @Risk model. You can add as many extra worksheets as you consider necessary, showing the results and analyses cut & pasted from @Risk. Please rename all these worksheets to make it clear what data they contain. You can refer to these in the written report.
Marking scheme (approximate)
@Risk model logic correct 30
Excel sheet laid out clearly 10
Use of sensitivity analysis 10
Presentation of results 20
Report for Manager – clarity of explanation, no jargon 20
Only relevant output attached (quality not quantity) 10
TSP and the oil platform contract
The construction company TSP has won a contract to build an off-shore oil drilling platform in the North Sea. It is a major project which also includes the development of land-based facilities to support the drilling operations. This involves building a small harbour, plus a number of other buildings including living accommodation, shops and offices, and improving the road link to the site. The project start date is May 1st 2015.
TSP is a relatively small and new company, and the success of this project is crucial to their future in a very competitive industry. TSP does not employ sufficient permanent staff to carry out the work, so the first task is to recruit contract staff for this project. They will need unskilled labour for the road improvements and for building the harbour, and specialised skilled personnel for a number of later tasks, including laying a pipeline from the platform to the shore and the installation of the drilling equipment on the platform.
The Human Resources department of TSP reckons that the time taken to recruit an unskilled labour force will be uniformly distributed between 30 and 60 days, and the time to recruit the specialised staff will be uniformly distributed between 60 and 120 days.
The time needed for the road improvements will depend on a number of factors such as the nature of the terrain and the reliability of the machinery used, but it is estimated that it can be modelled by a triangular distribution with min 50 days, most likely 90 days and max 150 days. Similarly, the time to build the harbour is also unpredictable, as it depends on geological factors, but can be modelled by a triangular distribution with min 100 days, most likely 120 days and max 200 days. Sufficient staff will be recruited to enable both these jobs to run concurrently.
As soon as the road improvements are complete, work can begin on erecting the buildings on the dockside. The time needed for this building work is fairly predictable and can be assumed to be 60 days.
When the harbour is ready, the platform itself can be built. This activity is highly dependent on the state of the weather; really bad weather could have a disastrous effect on the completion time. Estimates have been obtained for the probability of various weather conditions for the duration of this activity, and these are summarised in the following table.
Weather conditions OK Moderate Really bad
Probability 0.7 0.25 0.05
Time taken to build platform (days) 30 60 240
As soon as the platform is ready, the specialist staff can begin two activities, which can run concurrently. Firstly, they can erect the buildings on the platform. This is an activity of uncertain duration which can be modelled by a uniform distribution with min 50 days and max 80 days. Secondly, they can begin the much more complex task of laying the pipeline to the shore. The time needed to lay the pipeline depends on the time of year this activity can start, as it depends on the weather.
Start date Time to lay pipeline (in days)
1st May – 30th Sept
(summer) Triangular (30,60,120)
1st Oct – 30th April
(winter) Triangular (60,120,240)
As soon as the pipeline is in situ, the drilling equipment can be installed. This takes exactly 30 days.
TSP will be paid £35m if the project is completed by 30th April 2016 (i.e. within one year). However for each day by which the project is delayed, there is a penalty of £10,000, i.e. £10,000 per day will be subtracted from this payment.
TSP have two possible options for reducing the duration of this project. Firstly, they could contract out the building of the harbour to a local company. This would change the duration of this activity to a triangular PDF with min = 50, most likely = 80 and max = 120 days, but at a total cost of £200,000. Secondly, for a cost of £500,000, TSP could hire some special Norwegian equipment which would reduce the risk of a severe over-run for building the platform in the event of really bad weather. Instead of a worst-case of 240 days, this would be reduced to 150 days, with the other two possibilities remaining the same (30 days in OK weather, and 60 days in moderate weather).
Your task is to develop an @Risk model for this project and use it to write a report for the board of TSP on the risks involved in the project. In particular, you should provide the board with information to enable them to decide whether or not to contract out the harbour building and/or to rent the Norwegian equipment. You should also identify the areas of highest risk and suggest ways in which this might be managed.
Here is the activity diagram:
A Recruit unskilled labour
B Recruit skilled labour
C Improve road
D Build harbour
E Build dockside buildings
F Build platform
G Erect buildings on platform
H Lay pipeline
I Install drilling equipment
Modelling the “Lay pipeline” activity (H)
The choice of PDF for this activity depends on whether it starts in a summer or winter month. Obviously, this depends how long the preceding activities take. There are two paths which must both be complete before H can start: B and A-D-F. Activity B cannot take longer than 120 days but path A-D-F could potentially take more than one year, and thus there could be several changes between summer and winter before H can start. Thus the model needs to know whether the path A-D-F ends in a summer or winter month.
Begin by drawing up a table like this, which counts the number of days from the start of the project, converts this into months, and calculates the “changeover points” between summer and winter. Thus all days up to day 153 (Sept 30) are summer days, and day 154 (Oct 1) is a winter day.
Month Days in this month Cumulative number of days
May 2015 31 31
June 30 61
July 31 92
Aug 31 123
Sept 30 153
Oct 31 184
Nov 30 214
Dec 31 245
Jan 2016 31 276
.…. ….. …..
This is just the top part of the table – you will need more rows than this. To work out how many rows you actually need, look at the “worst case” i.e. the maximum possible length of path A-D-F.
Next you need a cell that calculates the max length of path A-D-F in days. Finally, you need to write a “nested IF” formula in Excel which basically says, if the start date falls in a summer month, then use RiskTriang(30,60,120), and if the start date falls in a winter month, then use RiskTriang(60,120,240). There is lots of help in Excel and on the internet (including YouTube videos) on using the IF function.
Suppose for the sake of argument that in one iteration the sampled length of path A-D-F was 206 days. In this case activity H could start in November, a winter month. However, if in the next iteration the sampled length of path A-D-F was 114 days, then activity H could start in August, a summer month.