ENERGY 722, 2017
ASSIGNMENT TRANSPORT & DECISION MAKING
Due Monday 22 May by 4pm.
Submit in electronic form via Canvas.
NOTE: Work on Task 1 now, wait until the end of Andrea’s lectures with Tasks 2 & 3.
Task 1 – Benefit Cost Analysis < 17 marks >
A road realignment project aims to identify a new route for an existing road that currently leads
through two towns. The existing road causes traffic congestion in the towns, which a new route
would alleviate. Five different alternative routes have been proposed. The alternative routes
were each assessed under the following five criteria:
Traffic benefits (in million $) made up of monetized travel time benefits, vehicle
operating cost savings and monetized accident savings.
Costs (in million $) are the construction costs and ongoing costs.
Noise level (in dBA), this is the maximum noise at least one resident adjacent to a
proposed route might experience. The noise level for residents adjacent to any of the
proposed new routes currently is no more than 40 dBA.
Effect on landscape is assessed on a categorical scale and ranges between insignificant,
minor, moderate, significant and severe.
Effect on ecology is assessed on a categorical scale and ranges between insignificant,
minor, moderate, significant and severe.
The different alternatives (proposed new routes for the road) were assessed resulting in the
following performance matrix, which shows alternatives in the columns and criteria in rows:
A B C D E
Traffic Benefits
Costs
Noise 50 55 60 55 65
Number of
households
affected by noise
50 100 60 80 40
Landscape Severe Minor Moderate Insignificant Moderate
Ecology Moderate Moderate Severe Minor Minor
1. Estimates of benefits and cost over a 10 year analysis period1 are provided in the
accompanying Excel spreadsheet. Derive the missing present value entries for both
benefits and costs for each alternative, and derive net present value (based on benefits
and cost) for the projects above. Assume a discount rate of 6%. <5>
2. The NZ Economic Evaluation Manual provides a standard monetary value for noise.
Costs of road noise are estimated to be
1 Note that the NZ Economic Evaluation Manual normally requires a 40 year analysis period.
$350 change number of households affected.
Determine the cost associated with the dis‐benefit of road noise, assuming for simplicity
the given formula already estimates the present value of noise dis‐benefit.
Derive net present value for projects A to E again, now also taking into account the disbenefit
of noise. <3>
3. Determine the benefit cost ratio of the different options here for all monetized criteria
(including noise). What are the three top‐ranked alternatives according to your
analysis? <3>
4. Would you recommend making special note of the other non‐monetised criteria
alongside your benefit‐cost analysis? Explain why or why not. <2>
5. Also compute the BCR of alternatives A to E now assuming benefits and costs are
discounted over a 7 year analysis period. Compare the results to those obtained with a 10
year analysis period, and explain what happens. <4>
For Task 1, submit a spreadsheet used for the calculations above together with a write‐up of
your analysis.
Task 2 – Value Functions in SMART Analysis < 18 marks >
Emma wants to buy a new e‐bike for her commute to work. Emma does not worry about the
cost of the bike. Her only criteria in this decision are how the bike looks, its battery capacity and
its weight. Battery capacity will be measured in Watt hours (Wh): the higher this figure, the
further the e‐bike will go ‐ a rough estimate is
, see 2. Since Emma wants
to store her bike in the basement she needs to be able to carry the bike down a few stairs, hence
weight (kg) is important. Emma measures the battery capacity of e‐bikes on a global scale
between 0Wh and 1000Wh, and weight on a scale from 0kg and 40kg. Bikes with higher
capacity batteries are generally heavier. Quality is assessed by direct rating.
Emma is considering the following bikes with specifications listed in the table. Emma has rated
the “look” of each bike out of 100 (where 100 is best):
Look Battery capacity (Wh) Weight (kg)
Alpha 420 27
Beta 480 22
Cruiser 620 24
Delta 350 20
a. Do you think Angela chose good limits for her scales? Explain briefly. <2>
b. Emma states that her value function for battery capacity is linear. The worst level of
attribute values on her scale is assigned a value of 0 and the best level is assigned a
value of 100. State the equation for this value function. <2>
2 Source: http://www.treehugger.com/bikes/should‐i‐buy‐electric‐bicycle‐everything‐you‐need‐toknow‐
primer‐faq.html
c. Emma derives the value function for weight based on the bisection method. She
determines that the midpoint value of weight is 15kg and the quarter points are at 10kg
and 25kg. Draw Emma’s value function. You do not need to state the equations. <3>
d. Using the swing weights method to derive the weights of the criteria she determines
that weight is the most important criterion. The second most important criterion is the
look of the bike where a swing in the look from worst to best is 70% as important as a
swing for the most important criterion. Battery capacity is the least important with a
swing from in battery capacity being 50% as important as a swing in weight. Derive the
normalised weights for all criteria. Show your working. <4>
e. Emma applies direct rating to assess the look of the bikes. She rates the look of the bikes
as Alpha: 75, Beta: 0, Cruiser: 80, Delta 40. Derive the value score of the battery capacity
based on the linear value function from part b, and the value score of weight from the
value function from part c (read the approximate values off the graph), and list all in a
table. <4>
f. Derive the overall scores of the bikes. Which one should Emma buy? Show your
working. <3>
Task 3 –SMART Analysis <37 marks>
Problem setup: Assume you are moving to New Zealand from overseas to work in Auckland for
3 years, and then move away from New Zealand. You will be working for a startup in the CBD, in
shared offices at the “Generator” https://generatornz.com at 22 Customs St E.
You have already found a place to live near the corner of Springleigh Ave and Woodward Rd in
Mt Albert, Auckland – assume you will be staying there for the whole period:
https://goo.gl/maps/XAcocmpLmWw
You will have to commute to work on a daily basis. Apply a SMART decision analysis to identify
what your preferred approach to travelling to work is.
1. State at least 5 unique alternative ways you will consider for getting to work. Make sure
they are well‐defined.
Examples: Your alternative could be to drive a car to the city and park it at one of the
commercial car parks there. Or your alternative could be to drive a car to somewhere
close to the city where you can park for free, but then walk to your office. Avoid
ambiguity and instead state clearly what you mean. Don’t forget to consider all costs
associated with a trip (including vehicle costs if applicable), and you may want to
include the return trip in your analysis.
Hand in: write at most ½ page. <5>
2. What are your objectives and criteria? To give you some ideas: objectives / criteria
could have to do with travel costs, time spent travelling, environmental impact, comfort,
reliability, and others. Describe the objectives / criteria you want to consider in your
analysis. Ultimately the (measurable and detailed) criteria should be grouped
underneath your major (general) objectives, and structured from general (top) to more
specific (bottom), similar to the value trees you have seen in class.
Note that your value tree should contain the most important criteria that you want to
include in your analysis, but not too many so that the following tasks do not get too
complicated. You need to have at least 5 criteria (or leaves) in your value tree. You can
use criteria and objectives listed above, and others that you find important. Make sure
that you briefly explain your criteria if they are not self‐explanatory.
To make sure you are not missing any important criteria think about what is particularly
good or bad about an alternative you listed in Task 3.1.
Hand in: 1 value tree, at most 1 page. <3>
3. For each criterion at a leaf of your value tree, indicate how you will measure it, and state
the units of measurement. Also give the performance matrix for each criterion you can
measure. (That means you may omit the criteria that you will assess by direct rating, but
an effort has to be made to measure all measurable criteria).
Examples: You could measure travel time by estimating daily commute time (in minutes).
In the performance matrix you then list your estimates of travel time. Another example
is comfort of travel. You may not be able to define a quantitative measure for comfort. In
this case you can state that you are assessing a criterion by direct rating, and leave the
corresponding entries in the performance matrix empty.
Briefly state your approach for obtaining estimates (e.g., for commute time by train I
would get an estimate from the Auckland Transport website, and then add an estimate
for walking and waiting time. Alternatively, google maps might give you good estimates
which also capture congestion at peak times). You need to find your own data here.
Some potential resources are listed below. Make sure you reference all data sources.
Hand in: Brief description of criteria measures, and approach ½ ‐1 page and performance
matrix. <8>
4. Derive the matrix of scores (0 – 100 with 100 being best). For those criteria you can
measure (Task 3.3), I suggest using a linear value function for simplicity. You also need
to assess the criteria you are scoring by direct rating and list the results in the matrix.
Briefly justify your direct ratings for one of your criteria. A sample Excel spread‐sheet is
provided on Canvas that should help with this step.
Hand in: matrix of scores, justification of direct rating explanation for one criterion. <5>
5. Specify the weights you choose for the different criteria, and briefly explain your choice
of weights. Remember the discussion we had in class on weights.
Hand in: List of weights for all criteria and brief justification, less than ½ page. <3>
6. Derive the overall SMART scores for each of your alternatives (Task 3.1) for the decision
problem given by your value tree in Task 3.2. Derive the scores based on the Excel sheet
provided. Briefly discuss whether the preferred alternative is the one you expected, and
justify your answer.
Hand in: SMART scores for each alternative, and brief discussion (1‐2 sentences). <5>
7. Conduct a brief sensitivity analysis. Describe how you conducted your sensitivity
analysis, and what your conclusions are. Also comment on what key uncertainties might
affect your decision.
Hand in: Brief discussion, less than ½ page. <4>
Note that some marks will be awarded for presentation of results <4>.
Hand in:
Hand in a summary of your work as outlined for each task above.
Also hand in the Excel sheet you used for your analysis.
Make sure you provide references to resources you use for the assignment.
Note that “the” correct answer doesn’t exist here. You need to justify and explain the
choices you make. You might each end up with a different decision model and answer.
Useful resources may be:
Google maps (good for travel time estimates, also taking into account traffic congestion
if you look at the corresponding time of day; public transport and walking; distance
measurement, etc.): https://www.google.co.nz/maps
Openstreemap showing cycling infrastructure:
http://www.openstreetmap.org/#map=13/‐36.8879/174.7422&layers=C
Openstreemap showing Public Transport lines:
http://www.openstreetmap.org/#map=14/‐36.8870/174.7188&layers=T
Auckland Transport website (has information on public transport – including costs,
cycling, walking, parking): https://at.govt.nz/
Cycle route that might be of interest in the area: https://at.govt.nz/cyclingwalking/
auckland‐cycle‐run‐walkway‐maps/northwestern‐cycle‐route/
Parking: http://www.wilsonparking.co.nz/go/regions/auckland‐cbd