Final Reseach Submission

European Hyperloop Week 2022

Team Name Shift Hyperloop

Point of contact and email Endre Vee Hagestuen, endre.hagestuen@shifthyperloop.com

Subcategory specification Feasibility studies and Predictions of socioeconomic effect

Award Showcase Demonstration Title of sub- System sub-

mission categories
Socioeconomic Yes No FRS Socioe- Route and
Aspects of conomic As- City Plan-
Hyperloop pects of Hy- ning
Development- perloop De-
Full Scale velopment
Award

June, 2022
Abstract

This report suggests a general route planning method, meant to optimize the route in terms of
cost and benefits. The possibility of hyperloop between the two most populated cities in Norway,
Oslo and Bergen along with a possible station location in Bergen is also investigated.
The general route planning method is based around a python script, found in Appendix referanse,
and can be implemented in other parts of the world with minimal adjustments. The main idea of
the method is to minimize the cost of construction, travel time and energy consumption.

The proposed station location is chosen based on the population density in the area, puplic trans-
port connections, availability of land and possible difficulties during the construction phase. Based
on these factors, the proposed station location in Bergen is located right next to the Bergen Railway
station.
Lastly, it is discussed how the hyperloop must be supported both economical and political support
to become a reality. Some suggestions for commercial funding and how to get political support is
also made.

i
Contents

1 General 1
1.1 Description of the participant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1.2 Development environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1.3 Research objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.4 Representative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1.5 Registered Design Competition Award . . . . . . . . . . . . . . . . . . . . . . . . . 1

2 Research 2

2.1 Motivation of the research project . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

2.2 Scope Of Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2.3 Route Planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

2.3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.3.2 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.3.3 Calculations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

2.3.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.3.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.4 City Planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

2.4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.4.2 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.4.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

2.4.4 Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.5 Political and Economical Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

References 13

Appendix 15
A Appendix 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
B Appendix 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

C Appendix 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

ii
1 General

1.1 Description of the participant

Shift Hyperloop is an independent non-profit organization founded in Trondheim by students from

the Norwegian University of Science and Technology (NTNU). The team currently consists of
approximately 60 members from various fields of study. An updated list of team members can be
found in Appendix A, in Table 2
Shift was officially founded 1. February 2019 by ten students with the goal of developing a green
and efficient transportation method based on Hyperloop technology. Shift Hyperloop enjoys a
close and mutually beneficial relationship with NTNU. Shift consists entirely of NTNU students
and gives them an opportunity to have a hands-on experience that is normally unavailable for
engineering students. New students get an excellent opportunity to meet others with similar
interests, work with more experienced students and build interdisciplinary relationships. We also
provide the university with potential bachelor- and master-theses for graduating students.
Shift Hyperloop is primarily an organization meant to teach and educate NTNU students. However,
it also consists of students who seek to help solve the massive challenges the world is facing. As
such, one can expect to see Shift advocating for green measures and further research on how to
combat climate change.

1.2 Development environment

Most work in Shift Hyperloop is done in our office close to our campus at NTNU. We have three
workshops per week, arranged in different groups. Each Sunday, we have a joint meeting with the
whole team, addressing issues that concern everybody. Parts of our design and production are
done with our sponsors’ help in the mechanical, electrical, and concept departments.
The Concept Group has workshops at the Shift office, and the work is done using several analogs
and digital creation techniques. For example, Google Earth Pro and Python have been used for
route and city planning. Other than that, maps of the geology of Norway have also been used.

1.3 Research objectives

This research includes finding the best possible route that connects Oslo to Bergen and possibly
also Oslo to Stavanger and finding the station’s location in both Bergen and Stavanger. The group
is also investigating the geography and geology in Norway to figure out where the route should go
to make it as efficient as possible. Furthermore, the research also includes analyzing the political
and economic aspects to determine how the Hyperloop can be made a reality in Norway.

1.4 Representative

Representative to be in correspondence with the EHW Committee:

Endre Vee Hagestuen, CTO Electrical, endre.hagestuen@shifthyperloop.com

1.5 Registered Design Competition Award

We want to emphasize the fact that this the Final research submission document. Shift Hyperloop
is participating in the full scale award: ”Socioeconomic Aspects of Hyperloop in Norway”.

1
2 Research

The concept group in Shift Hyperloop Team 2021 investigated where the most profitable and
efficient routes would be for hyperloop in Norway. This resulted in having the central station in
Oslo that would further go to Stavanger, Bergen, and Trondheim. In addition, the route from
Oslo to Trondheim was further investigated. The main hyperloop station was placed in Oslo,
underground at ”Christian Frederiks plass” close to Oslo City. The thought process behind this
decision is well explained in last year’s Full-Scale Award ([19]). This location is further used in
this study to plan the route connecting Oslo to Bergen.

This year’s Route and City planning group researched the possibility of a general method finding
the most profitable and efficient Hyperloop route, and the economic and political challenges/aspects
the implementation of Hyperloop in Norway can encounter.
Using the general method, the team planned a route between Oslo and Bergen, with a station in
both Oslo and Bergen. These cities are decided to be studied further due to being the most promi-
nent cities in Norway, making them more likely to be economically and politically favorable, even
with sub-optimal geography. Furthermore, many mountains characterize the geography between
Oslo and Bergen. Due to this, the route is not the easiest to build, but still possible when building
tunnels. The goal is to find the optimal route regarding build and operation costs and discuss how
this could get political and economic support, making it a reality.

Furthermore, the ITSR explained that the route between Oslo and Stavanger would be investigated
further if there was time. Unfortunately, this was not the case, and the route from Oslo to Stavanger
was not planned. Instead, the focus has been on developing a general method in python for a
potential hyperloop route.

2.1 Motivation of the research project

The motivation behind the research is to investigate how a futuristic transportation method such as
the hyperloop can be possible in Norway. Future transport methods need to be less polluting, and
the hyperloop can be a solution here. This makes it so interesting to investigate the implementation
of the hyperloop and the political and economic aspects that will follow.
When it comes to route planning, the team believe a general method for every possible route could
be made. By creating a general method, later routes could be planned a lot faster and hopefully
better. The goal was by no means to create a flawless method during this year, but rather lay a
foundation other research could build on and eventually making the method close to perfect. The
team believes that a perfected general route planning method would save a lot of future planning
time and also result in a better route than humans could ever plan manually.

2.2 Scope Of Research

The scope of the research includes finding a general method for semi-automatic route planning.
Furthermore, the scope also includes finding a station location in Bergen and analyzing the political
and economic aspects of building a hyperloop.

2.3 Route Planning

This section will present the research done by the Concept group on planning a future hyperloop
route in Norway. The routes investigated are based on last year’s work.

2
2.3.1 Introduction

Due to Norway’s hilly nature, the hyperloop seems less effective in Norway than in other countries.
Therefore, the concept team finds it more valuable to research the possibility of a general method to
decide on a potential hyperloop route in Norway rather than planning a specific route. However, the
route between Oslo and Bergen will be examined and is an example of how the method developed
can be used to define a potential route. Furthermore, it can be utilized for routes in other parts
of the world with minimal adjustment.
A potential hyperloop route is presented by first plotting waypoints along the route in Google
Earth Pro and then saving a list of the waypoints along the route using GPS Visualizer. The next
step is using a self-written Python script, linked in Appendix C, to complete essential calculations,
thereby giving the relative cost of construction and operating the hyperloop route. The parameters
considered are the cost of construction both the hyperloop and potential tunnels, estimated travel
time, and estimated energy use.
The construction costs are one-time costs if maintenance of the track and pods are neglected. The
maintenance cost is assumed to not differ significantly depending on the route between the same
cities and is therefore neglected for simplicity. On the other hand, the running cost is continuous,
meaning the relative cost of different routes will vary depending on how many times the route is
traveled.
The team, therefore, optimized the route’s value after 50 years, with ten pods running at once for
12 hours each day. Unfortunately, there seems to be very little research done on hyperloop lifetime,
but the lifetime of a steel railway is estimated to be around 30 years ([10], p. 17). Furthermore,
the hyperloop avoids a lot of the wear and tear railway experiences due to the protection from
the tube and the magnetic levitation. Therefore, the lifetime might be over 50 years, making it a
reasonable time after build to optimize value. However, this research is insufficient and thus should
be examined further.

To transport as many people as possible without any logistical problems, the team is looking into
having ten pods running at once. This makes the time difference between each pod 5 minutes,
meaning the time for entry and disembarkation 5 minutes. Therefore, the hyperloop can transport
several people equalling the pod capacity every 5 minutes, in both directions, when active. In this
case, the pod capacity is 14 passengers ([20]). Therefore, if the pod is operational 12 hours a day,
the total number of transported passengers each day will be 14 ∗ 12∗60 5 = 2016 passengers every
day in each direction.
According to SSB, the total number of people transported in the first quarter of 2022 on domestic
flights in Norway was 2 785 968; from this data, 884 460 flights were from and 404 206 from Bergen
([21]). Using these numbers, a good approximation for how many people transported from Oslo
to Bergen by plane would be 884 460 ∗ 2 785 404 206
968−884 460 = 188 011 passengers this quarter. With
2016 transported passengers each day, the suggested hyperloop can transport 2016 ∗ 90 = 181 440.
These numbers are similar, showing how the hyperloop can practically remove the need for planes
in terms of passenger transportation.

2.3.2 Methodology

As mentioned in 2.3.1, Google Earth Pro was utilized to plot the routes and a self-written Python
script to make necessary calculations needed to find the relative cost of the route after 50 years.
This allows examining multiple routes quickly and therefore increases the chances of finding the
best possible hyperloop route based on the boundaries set by the team.
Firstly, the tool “path” is used in Google Earth Pro to draw a route from the hyperloop station
in Oslo to the hyperloop station in Bergen. When the path is drawn, multiple waypoints appear
along the path. The waypoints are used to calculate the curve radius and velocity and to estimate
the tunnel length needed. After plotting the route, the elevation profile opens, and one can see the
total length of the route.

3
When the path is completed, it is saved as a kml-file, and uploaded to a GPS Visualizer website,
which converts the kml-file to a text file. This provides a table of waypoint coordinates with the
belonging elevation, which is read in the python script. Unfortunately, the total length of the route
is nowhere to be found in the text file, meaning the route length needs to be inputted manually,
using the value found in Google Earth Pro. However, the python script creates a list of waypoint
coordinates transformed to the correct UTM-zone, in our case, UTM 32V. In UTM coordinates,
the x- and y-axis are in meters ([13]) , and it is necessary to calculate the curve radiuses in meters
along the route.
The curve radiuses are calculated by using the equation for a circle (x−x0 )2 +(y −y0 )2 = r2 , where
the center is defined as (x0 , y0 ), and placing three neighboring waypoint coordinates (x,y) on the
perimeter of the circle. By setting a limit for maximum centripetal acceleration ac,max , one can find
2
the maximum speed the pod can maintain using the formula for circular motion ac,max = vr . No
matter how big the radius is,a limiting speed of 1200 km/h is set as this is just below the maximum
speed, according to Elon Musk’s Hyperloop Alpha paper ([14], p. 12). From the velocities, the
speed the hyperloop can maintain at a certain point of the route is plotted, together with a 10th
polynomial regression model. The 10th polynomial model is used in later calculations, as the team
believe this is a better and smoother representation of the speed development.
After plotting multiple possible routes, they are compared by their relative cost, and the best
route plotted is the one with the lowest relative cost. How this relative cost is found is described
in Section 2.3.3.

2.3.3 Calculations

In this section the calculations provided in the python script will be described. Hence, it will give
more detailed information on the calculated relative costs, velocity and acceleration, construction
and pod costs, tunnel costs, and time and energy costs.
Relative Cost
The relative cost approximates the cost of constructing and operating the hyperloop for 50 years.
During the 50 years, there are costs for every trip, such as energy consumption and value of time
for the passengers. The approximation used to estimate operating cost is shown in Equation 1.
The operatingCost is defined as the relative cost related to each trip. timeCost is the value of
travel time.

operatingCost = timeCost ∗ numberOf P assengers + energyCost (1)

The one-time costs that have considered are the cost of construction of the hyperloop itself, po-
tential tunnels, and pods. The equation used to estimate fixed costs is shown in Equation 2.

OneT imeCost = constructionCost + tunnelCost + numberOf P ods ∗ podCost (2)

The relative costs are the sum of the one-time and operating costs, where the operating cost is
multiplied by the number of trips each way during the 50 years. This number of trips is in this case
numberOf T rips = numberOf OperatingDays ∗ activeHoursEachDay
travelT ime = 50 ∗ 365 ∗ 12∗60
25 = 525 600.
An adjustment factor is used to avoid unnecessary large numbers for easier comparison. In this
case, this factor is a = 10−6 . The total score is shown in Equation 3.

relativeCost = a ∗ (OneT imeCost + numberOf T rips ∗ operatingCost) (3)

Further calculations of each variable used in Equation 1, 2 and 3 is described in the following
paragraphs.

4
Velocity and Acceleration
Several of the later calculations are based around the movement of the pod, it is therefore necessary
to find information about the velocity and acceleration first. From the circular motion formula
√
introduced in Section 2.3.2, the expression for the operating speed is v = ac,max ∗ r. The maxi-
mum centripetal acceleration ac,max is therefore needed. The velocity between the start and stop
phase is meant by operating speed.

The ITSR used a centripetal acceleration limit of 0.1g due to the passenger’s comfort. After
further research, there appears to be some disagreement around this limit. The Hyperloop One
paper operates with a maximum centripetal acceleration of 0.5g ([14], p. 40). The team decided
to operate with a total acceleration, both centripetal ac,max and acceleration due to speed change
av,max combined, limited at 0.5g. For simplicity, the limit for both ac and av will be costant.
q
Using the Pythagorean theorem, one gets have the equation av,max = 0.25g 2 − a2c,max . During
the start and stop phase, av = av,max to reach either operation speed or full stop as quickly
as possible. The speed still varies at operating speed between the start and stop phase. This
acceleration will, however, never be close to nor exceed av,max , meaning the operation speed is
only determined by ac,max . A specific operating speed is often simplified to the average operation
speed vavg . The longer the pod can maintain operation speed, the shorter the travel time. On the
other hand, the higher av,max is, the lower ac,max and the operation speed is. To lower the travel
time, the perfect relationship between ac,max and av,max should be found.

During the acceleration at the start to operation speed, the average speed is half the operation
speed since the acceleration av = av,max is constant. This means that the time lost by accelerating
compared to starting at operation speed is half the time used to accelerate to an operation speed.
The exact time loss will happen when braking, meaning the total time loss is the same as the time
vavg
used to accelerate tloss = av,max .

By manually adjusting ac,max for a good route, one could see how the score changed and used
the ac,max that gave the best score. √
That resulted in ac,max = 0.32g and thereby av,max = 0.38g,
meaning the operation speed is v = 0.32 ∗ g ∗ r.
Construction- and Pod Costs

In construction costs, the construction and groundwork costs is included. Depending on the
bedrock, the cost of groundwork will vary from location to location. The hyperloop will pre-
sumably require crossing the same bedrock no matter the route, causing the difference in cost
between routes to be negligible. Therefore, it was decided to use a constant build cost per kilo-
meter, the same as described in the Hyperloop Alpha paper. This paper estimates the total cost
of construction without tunnels to be 4810 million USD ([14], p. 56). The route length is 563 km
([14], p. 26), resulting in a cost of 4810
563 = 8.54 million USD per km. The construction cost in USD
used in Equation 2 is therefore constructionCost = 8.54 ∗ 106 ∗ length.
In the Hyperloop Alpha paper, a total pod cost of 54 million USD is estimated for 40 pods ([14],
54
p. 56) carrying 28 passengers each ([14], p. 9), meaning 40 = 1.35 million USD per pod. As
the pod only carries 14 passengers, it is reasonable to assume somewhat lower cost. The estimate
is podCost = 1 ∗ 106 USD, used in Equation 2. However, the pod cost does not affect the route
choice, as this cost is the same no matter the route.
Tunnel Costs
Due to the high speed, the hyperloop cannot follow hilly grounds. The limit for curvature in the
vertical plane is even stricter than in the horizontal plane. According to Railovution, a vertical
radius of 118km is necessary for a speed of 1200km/t ([16]). This is virtually a straight horizontal
line. The elevation differences of the ground can be somewhat neutralized by adjusting the length
of the pylons supporting the tube. However, the pylons can only adjust for elevation differences in
slope gradients up to 6%, according to the Hyperloop Alpha paper ([14], p. 42). In terrain steeper
than this, the tunnel is needed.
To find areas where the slope is steeper than 6%, the elevation differences ∆h between waypoints is

5
used. For further calculations, a segment is defined as the part of the route between two waypoints.
By assuming the distance between each waypoint is equal throughout the route, the length of each
segment l can be found using the total route length and the number of segments n (which is known),
totalRouteLength ∆h
l = numberOf Segments . The average slope gradient a is defined as a = tan(θ) ∗ 100 = l ∗ 100. For
each segment between two waypoints where the average slope gradient is more than 6%, theP total
n
tunnel length increases with the distance l. This results in en expression tunnelLength = l ∗ i=1
(segmenti WHERE a > 6%).
According to Hyperloop Alpha Paper, the tunneling cost per kilometer is estimated to be around
31 million USD per kilometer ([14], p. 42). This results in a tunnel cost of tunnelCost =
31 ∗ 106 ∗ tunnelLength, used in Equation 2.
Time Costs
When planning the hyperloop route, it essential to consider the travel time. Reducing travel time
makes it reasonable to increase the ticket price as the passengers save time. The travel time is
found from the average operation speed, total route length and the additional time loss due to
start and stop, giving travelT ime = length
vavg + tloss . The value of travel time savings (VTTS) seems
to be around 14.8 CHH/h ([18], p. 186), which translates to about 15.24 USD/h. The time cost,
used in Equation 1 is therefore timeCost = V T T S ∗ travelT ime.
Energy cost
Finding the energy consumption, one firstly needs an expression for the acceleration. This means
an expression of the velocity v as a function of the time t, instead of the distance x. To do this, the
x-axis was scaled according to the travel time and total route length to have a simplified function
v(t).
In this part, a segment is defined as a part of the route with velocity increase. For each segment
v(t2 )−v(t1 )
of velocity increase, the average acceleration will be aavg,segment = ∆v ∆t = t2 −t1 . The t-
d
values represented by t1 and t2 are found by solving dt v(t) = 0 and v(t2 ) > v(t1 ). The average
acceleration for each segment is added to aPsum and then divided by the travel time to find
n
(a ∗tsegmenti )
total average positive acceleration: aavg = i=1 avg,segment i
travelT ime . Negative acceleration is
neglected in terms of energy consumption.

Using the average acceleration, the energy consumption will be E = aavg ∗ mass ∗ length. The
mass used in this case is 3000kg, the pod weight is estimated to be 2000kg ([20]), and with 14
people on average weighing 71kg. According to Statista, the average price per kWh was 0.1118
0.1118 −8
USD in 2021 ([1]), or energyP rice = 3.6∗10 6 = 3.11 ∗ 10 ) USD per Joule. The energy cost one
way, used in Equation 2 will therefore be energyCost = E ∗ energyP rice.

2.3.4 Results

Last year’s Shift hyperloop team planned a route from Oslo to Trondheim, exiting Oslo eastwards
([21). To reuse parts of this route on the route to Bergen, the team examined routes exiting Oslo
similarly. Also, in the ITSR, an area called Hardangervidda, shown in Appendix B, Figure 6, is
protected, where building a hyperloop will be prohibited. Due to this, it would not be possible to
have a route in a straight line.

6
After plotting numerous similarly good routes, the route with the lowest relative cost was the
following route from Oslo to Bergen, shown as red in Figure 1. The kml-file of the routes plotted
is found in Appendix C.

Figure 1: Best plotted route (red) next to other plotted routes (blue).

With a route length of 465 km, and an average speed of 1165 km/h resulting in a travel time of 25
min 23 seconds, this route had the lowest relative score out of the ones plotted. As a result, the
estimated energy consumption and tunnel length are 150 kWh one way and 42 km, respectively.
This information and the relative cost are shown in Figure 2.

Figure 2: Output from Python script.

For comparison, the output from the routes plotted in blue in Figure 1 is shown in Figure 3. None
of these routes have a relative cost close to the route already presented.

Figure 3: Output from Python script of other plotted routes.

7
Figure 4 shows the velocity the pod can maintain at certain points of the route, described by
the ”accurate” graph which found directly from the route’s waypoints, and a 10th polynomial
regression model of the same data. As seen, the pod reaches operating speed very soon and is able
to maintain a pretty constant and high operating speed.

Figure 4: Best plotted route (red) next to other plotted routes (blue).

2.3.5 Discussion

The main idea of this route planning method is to make it as general as possible and let the
computer do the specific route planning work. The method does this to some extent, although it
still lacks accuracy. Through further development however, we believe this method could become
more accurate, and eventually do a much better route planning job than a human could ever do
manually. A great general method will reduce the cost of construction and operating the route
significantly.
To improve this method, future research should focus on finding more accurate values for the costs
such as hyperloop build cost and tunnel build cost. This has not been a priority this year, and the
estimates seem to be uncertain. It is also worth noting that inflation is not taken into consideration
in the previous calculations. Furthermore, there are some specific areas we feel the route planning
method could be improved; these are discussed below. Most of these do not seem to affect route
choice by a lot, as the relative cost will remain more or less the same, but improving them will
make the total cost estimate more accurate.
Route plotting

When plotting the route, it is worth noting that more waypoints will result in a more detailed
route. With too many waypoints however, one might get sharp turns resulting in an unsteady
radius and speed. We tried plotting the routes with as many waypoints as possible, while keeping
the route smooth. This seemed to be very hard, as the speed graph, shown in Figure 4, showed a
somewhat unsteady speed. Some sort of automatic plotting would therefore be advisable.

For the speed distribution v(x) along the route to be correct, we assume an identical distance
between each coordinate. This is a little inaccurate as it was done by hand, but the error will most
likely not affect the further calculations significantly. It would however once again be preferred to
have some sort of automatic plotting.

Energy consumption
To find the energy consumption, an expression for the acceleration a(t) and therefore the velocity

8
v(t) is needed. In the method, we scaled the x-axis according to the travel time and total route
length to find v(t) from v(x). This is unfortunately just a simplification of reality. Since the speed
varies, the relation between distance x and time t is not linear, thus not scalable with a constant.
Looking at the speed distribution v(x) in Figure 4, the operation speed is somewhat constant
indicating that this simplification is not way off.

Both aerodynamic and magnetic drag forces are neglected in the energy consumption estimate.
This means that the energy consumption will be higher in reality. As aerodynamic and magnetic
drag forces vary depending on aeroshell design and levitation system, respectively, we believe it
is less relevant for route planning. The pod’s velocity does however impact drag forces, meaning
that the drag forces might impact optimal route choice. This means that it will be impossible
to calculate an accurate estimate for energy consumption, without knowing more about the pod.
Cooperation between the route planning group and the technical groups is therefore important to
find the these drag forces’ impact on energy consumption.
Relation between ac and av
In the method, the optimal ac,max was found for one specific route, and this value was used for other
routes as well. The optimal ac,max might differ slightly for other routes between the same locations
but will not affect the score significantly and likely not change the route choice. However, if two
routes have very similar scores, it would be relevant to find optimal ac,max for each route separately
using the method described under 2.3.2. In further route planning, one could investigate creating
a method for calculating the optimal ac,max automatically to make this process more efficient.

So far, we have just thought of ac,max and av,max as constant limits. A better solution could
however be found if the limits vary depending on the accelerations at the specific time. For
example, in the middle of the route the acceleration in direction of travel av is nowhere near the
limit av,max . This means that the limit for centripetal acceleration ac,max could be even higher at
this point while still having a total acceleration a < 0.5g. To do this, an accurate function of the
acceleration a is needed, which means an accurate function of the velocity v(t) is needed. As of
now, the function v(t) is simplified, v(t) needs to be improved before variable acceleration limits
are implemented.

2.4 City Planning

This section will present the research done by the Concept group on planning a future hyperloop
station in Bergen, Norway.

2.4.1 Introduction

The concept team of 2021 did the work on the placement of the station in Oslo. Factors considered
before deciding on the station’s location were public transport connection, availability of land to
build the station, the number of people living within a 2 km radius of the station, accessibility
for the big machinery during the construction phase, and geographical ground conditions. After
these factors were considered, Christian Frederiks plass near Oslo S was the perfect place for the
hyperloop station. More on this can be found in last year’s Full-Scale Award: ([19]).

The concept group 2022 worked on finding the best place for the station in Bergen. However, since
the route between Oslo and Stavanger was not further investigated, the group did not research to
find the station’s location at Stavanger.

2.4.2 Methodology

The factors considered for the hyperloop station location were the number of people living within a
10 km radius, public transport connections, availability of land for building, and possible disruption
during the construction phase. This year, the number of people living in a 10 km radius is used

9
because Bergen is less densely populated than Oslo. In addition, two locations were considered
for further research. The first option was to build a hyperloop station integrated with the train
station. The second option was near the Bergen airport.
Bergen has more space to build the hyperloop stations compared to Oslo. Both locations considered
in Bergen are viable options for a future hyperloop station due to the spacious area. As there is
space to build stations over the ground in Bergen, a station under the ground was not considered.
The total population of Bergen is 265 470 ([22]), and with almost 60% of the people living in Bergen
live in and around the city center ([3]), it would be ideal to have the hyperloop station close to
the city centre as well. It takes around 24 minutes to drive and about 50 minutes by ”Bybanen”
(light railway) from the city center to the airport, making a hyperloop station by Bergen Airport
far from where the majority of people live. Bergen airport does not have a connection with the
national rail either, so it would not be an ideal station location.

2.4.3 Results

Several factors including connections with other forms of transport makes Bergen train station
most suitable location for hyperloop station. The train station is in the city center and has good
connections with local bus and light rail services.

The proposed hyperloop station by Bergen Railway Station is shown in Figure 5.

Figure 5: Location of proposed hyperloop station at Bergen Station.

The area around the station sits on the bedrocks metagabbro, amphibolite, eclogite, peridotite,
quartzite, mica gneiss and aluminosilicate gneiss. These bedrocks are about 1460-1000 million
years old and are favourable for building the station on ([15]). In the conversation with professor
Krishna Panthi, the group members were advised that these rocks do not require special treatments
or other measures before the building can be started.
If the station is placed next to the train station, most of the people living in and around Bergen
can reach the station within 30 minutes by cycling, walking or driving. The train station itself
is a protected structure since 2003 and no changes can be made to the station ([5]). There is an
available space next to Bergen train station to build a hyperloop station as shown in Figure 5.

10
2.4.4 Discussions

The number of people living in proximity and the connection to another form of transport made
the Bergen rail station the most favorable option for building the hyperloop station. Furthermore,
the integration of the hyperloop station with the rail station and the proximity of bus stations
and the light rail make the hyperloop portal the central hub at Bergen. Furthermore, the station’s
location at Bergen allows the hyperloop to be out of the city as soon as it leaves the station, with
minimal disruption to the people and businesses around, and extensive city planning is avoided.
The hyperloop will take a similar route out of Bergen as the national rail. Placing the station near
the airport would have resulted in investement in connecting transport to the city center which
would add to the cost of already high hoperloop investment.

2.5 Political and Economical Aspects

Building the hyperloop network is not cheap. Elon Musk predicted in 2013 that the hyperloop
would cost around 8.5 million dollars (about 100 million NOK) per kilometer ([14], p. 26 and
p 56). Some studies show that the cost of building a hyperloop will be much higher than Musk
predicted ([2]). Furthermore, the leaked documents obtained by Forbes show that Hyperloop One
— one of the companies attempting to make the hyperloop a reality — is estimating the cost of
building the hyperloop to be around 5 million USD per Km. ([2]). Since there is no built hyperloop
project to compare the cost, it is not easy to estimate the cost of building the hyperloop network
in Norway. To analyze the political and economic aspects, we took a cost of 8.5 million dollars per
kilometer, as suggested by Elon Musk.
Despite the high cost, the opportunity the hyperloop brings is also immense. The investments
we make today do not necessarily give direct economic benefits in the short term and may look
like a waste of money without benefits, as some critics consider the hyperloop to be. However,
social, political, and environmental benefits should be considered when examining any project’s
profitability. This information is taken into consideration when investigating the economic viability
of the Hyperloop.
The route proposed by the team between Oslo and Bergen is about 465km. The minimum distance
of the route that connects Oslo, Bergen, Trondheim, and Stavanger would be around a minimum of
1500 km (straight line distance). If we take the minimum cost of building a hyperloop as suggested
by Elon Musk in 2008, the minimum cost to build a hyperloop network in Norway that connects
four cities will be above 12.75 billion USD. The national budget for transport sectors which includes
roads, railroads, airports, and other transport structures between 2022-2033, is NOK 1,076 billion
(about 100 billion USD). The Norwegian government allocated 32.1 billion NOK ($US 3.51 billion
USD) towards investment in railway infrastructure projects, operation, and renewal in 2021 ([6]).
The hyperloop, with a price tag of 2.75 billion USD, will not be a reality in Norway without public
investment and support, but the budget for the next ten years shows that there is no money to
build it entirely with public investment. There need to be other sources of funding to make the
hyperloop a reality in Norway. As with other infrastructure projects in Norway, the group expects
a potential hyperloop system will be financed primarily by commercial debt. Capital markets
are likely to represent a significant source of funding for hyperloop in Norway. Other sources of
financing, such as construction finance and infrastructure funds, are likely to be available.
Bane NOR SF is a Norwegian state enterprise responsible for railway infrastructure in Norway.
Bane NOR’s “Norwegian High-Speed Rail Assessment 2010-2012” identifies potential sources of
funding for high-speed rail in Norway. The group discussed the various funding sources Bane
NOR suggested in 2012 and found out many of those funding sources are relevant for the potential
hyperloop route. Table 1 shows the potential source of funding for the hyperloop system. ([8]).
Furthermore, PPE (public-private partnership) model suits best for the financing of such large
infrastructure projects.

Political support for any infrastructure project is equally important. Across Europe, there are a few
examples like HS2 in the UK, where politics has brought the project to a standstill, and projects

11
 Table 1: How to potential source funding of Hyperloop

Direct or Indirect government Funding Commercial Funding Time

Direct Government Funding Societe General
Nordic Investment Bank DNB
European Investment Bank Nordea
Trans European Transport Network BNP Paribas
CEF Programme —

like the channel tunnel that connects the UK and France, which studies show will never break even,
is a reality because of political will. Further, studies suggest that rail services between Tromsø
and Fauske in Norway are not economically viable but will probably be built in the near future
because of politics. The group contacted the nine political parties and asked about their position
on developing the hyperloop in Norway but unfortunately, no answer was received from any of
them. Just one political party, Miljøpartiet De Grønne(MDG), in Norway has known support for
hyperloop ([23]). This may be due to political parties donot have specific opinion on hyperloop
itself.
Norway competes with countries like China, South Korea, and Italy to build a floating tunnel
([11]). If the hyperloop could be integrated with the project to build the floating tunnel (hyperloop
through the floating tunnel), it would be easier to get public and political support for the hyperloop
project.
A study forecasts that a hyperloop planned in Canada will add 19.2 billion USD to the region’s
GDB, creating 140 000 jobs ([17]). According to the most recent data from the world bank, Norway
had a GDP of 362 billion USD in 2020 ([24]). A similar benefit to the economy as the project
in Canada makes a considerable difference to the county’s economy. Since Norway does not have
a problem with joblessness, the extra job created can also solve the job problem in neighboring
countries like Spain and portugal .
Norway is the EU’s 8th most important partner for trade in goods, With almost 95 billion USD of
trade between the EU and Norway ([4]). Fresh fish is one of the biggest export of Norway. Most of
these goods are shipped in the container ship. Container ship shipment is one of the slowest modes
of shipment and one of the most polluting means of transport with an adverse effect on marine
life. ([7]). The world is dealing with the supply chain issues, so the Hyperloop networks between
cities in Europe can be a green and quick alternative to the container ships and solve the supply
chain issue, which is forecasted to get worse in the future ([9]). Virgin hyperloop is one of the
first companies to recognize the hyperloop’s opportunities in the shipping industry. The company
recently said the company is in discussions with airports and port facilities worldwide to create a
pilot program for cargo shipment services, said Ryan Kelly, Virgin Hyperloop’s vice president of
marketing and communications ([12]).
The recent war in Ukraine has shown that Europe urgently needs to reduce the consumption
of fossil fuels. Moreover, have a plan for transportation of people and goods in case of similar
situation comes to other counties. The hyperloop is the best place to do that job because of its
speed and non-reliance on fossil fuels, and if the people are made aware of the hyperloop’s role in
such situations, the support for the hyperloop among people will grow further.

12
References

[1] Alves, B. (8. April 2022). Average retail electricity prices in the United States from 1990
to 2021. Statista. Obtained from https://www.statista.com/statistics/183700/us-average-retail-
electricity-price-since-1990/
[2] Bhuiyan, J. (26. October 2016). The Hyperloop will take a lot more money to build than Elon
Musk anticipated. Vox. Obtained from https://www.vox.com/2016/10/26/13425592/hyperloop-
one-elon-musk-cost-leaked-documents
[3] Cybo (unknown date). 190 Postal Codes in Bergen. Obtained from https://postal-
codes.cybo.com/norway/bergen/
[4] Eurostat (February 2022). Norway-EU – international trade in goods statistics. Ob-
tained from https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Norway-EU -
international trade in goods statistics
[5] Friele jr., O. (12. March 2003). Bergen stasjon fredet. NRK. Obtained from
https://www.nrk.no/vestland/bergen-stasjon-fredet-1.198209
[6] Hareide, K. A. (25. June 2021). National Transport Plan 2022-2033. Regjerin-
gen. Obtained from https://www.regjeringen.no/en/dokumenter/national-transport-plan-2022-
2033/id2863430/
[7] Haskell, D. G. (12. April 2022). An ocean of noise: how sonic pollution is hurting marine life.
The Guardian. Obtained from https://www.theguardian.com/environment/2022/apr/12/ocean-
of-noise-sonic-pollution-hurting-marine-life
[8] Jernbaneverket. (26. May 2011). Summary of Phase 2
Works. Norwegian High Speed Rail Assessment. Obtained from
https://www.banenor.no/contentassets/eeab2b5193e24df2a4f71c3db9604a91/nhs-ph2-
summary-report v02.pdf.
[9] Knight, W. (28. March 2022). The Supply Chain Crisis Is About To Get A Lot Worse. Wired.
Obtained from https://www.wired.com/story/supply-chain-crisis-data/
[10] Leong, J. Murray, M. H. (February 2008). Probabilistic analysis of train-
track vertical impact forces. Institution of Civil Engineers. Obtained from
https://www.icevirtuallibrary.com/doi/epdf/10.1680/tran.2008.161.1.15
[11] Lo, A. (29. January 2019). Can Norway win the global race to build a ‘floating tunnel’ ?.
CNN. Obtained from https://edition.cnn.com/style/article/norway-underwater-floating-tunnel-
intl/index.html
[12] Mahoney, N. (2. February 2022). Virgin Hyperloop wants to revolutionize freight transport.
Freightwaves. Obtained from https://www.freightwaves.com/news/virgin-hyperloop-wants-to-
revolutionize-freight-transport
[13] MassGIS, (1. May 1996). MassGIS Data: UTM Grid and Points. Mass.gov. Obtained from
https://www.mass.gov/info-details/massgis-data-utm-grid-and-points
[14] Musk, E. (2013). Hyperloop Alpha. SpaceX. Obtained from
https://www.tesla.com/sites/default/files/blog images/hyperloop-alpha.pdf
[15] Norges Geologiske undersøkelse. (2. August 2021). Berggrunn. Obtained from
https://www.ngu.no/emne/berggrunn
[16] Railovution (4. March 2020). Hyperloop – The Train-In-A-Vacuum-Tube Fantasy. Obtained
from https://www.railvolution.net/news/hyperloop-the-train-in-a-vacuum-tube-fantasy-5
[17] Railway Technology (30. March 2022). TransPod receives $550m for Edmonton-Calgary hy-
perloop system. Obtained from https://www.railway-technology.com/news/transpod-hyperloop-
system/

13
[18] Schmid, B. et al. (15. June 2021). The value of travel time savings and the value of
leisure in Zürich: Estimation, decomposition and policy implications. Elsevier. Obtained from
https://www.sciencedirect.com/science/article/pii/S0965856421001658
[19] Shift Hyperloop Concept Group (April 2021). Full Scale Award. EHW. Obtained from
https://shifthyperloop01.it.ntnu.no/confluence/display/ST2/Concept?preview=/10586459/25231544/ExtendedAbs

[20] Shift Hyperloop Concept Group (June, 2022). Final Research Submission. EHW. Obtained
from https://drive.google.com/drive/folders/1COq8fvg2wXPRV8xTnNljeYeiCJEt Df
[21] Statistisk sentralbyrå (26. April 2022). Air transport. Obtained from
https://www.ssb.no/en/transport-og-reiseliv/luftfart/statistikk/lufttransport
[22] Statistisk sentralbyrå (26. October 2021). Population and land area in urban settlements.
Obtained from https://www.ssb.no/en/befolkning/folketall/statistikk/tettsteders-befolkning-
og-areal
[23] The Local (22. May 2017). Norwegians want futuristic vacuum train between Oslo and Copen-
hagen. Obtained from https://www.thelocal.no/20170522/norwegians-want-futuristic-vacuum-
train-between-oslo-and-copenhagen/

[24] The World Bank (2020). GDP (current US$) – Norway. Obtained from
https://data.worldbank.org/indicator/NY.GDP.MKTP.CD?locations=NO

14
Appendix

A Appendix 1

Name Group Role

Hannah Ruyter Management CEO
Erik Soltveit Stange Management CFO
Magnus Gjerstad Management CTO Mechanical
Endre Vee Hagestuen Management CTO Electrical
Mathias Borge Kristiansen Management CCO
Konrad Sandtrø Management CMO
Gonzalo Sanfeliu Moreno Management Systems Engineer
Ian McGregor Aksum Relations Group leader Relations
Daniel René Astor Relations Marketing Coordinator
Henrik Laukli Relations Marketing Coordinator
Mikkel Dahl Slettebø Relations Web Developer
Eik Hvattum Røgeberg Relations Web Developer
Jonas Bakke Powertrain Group leader Powertrain
Marius Tomasgård Bjørgan Powertrain Power Electronics
Daniel Stjernen Powertrain Power Electronics
Audun Bangsund Powertrain Power Electronics
Steve Carter Fuejo Nomeny Powertrain Motor Control System
Håvard Sinnes Haughom Powertrain Motor Control System
Jørn Andre Gundersen Powertrain Motor Design and Simulation
Ivar Beseth Nordeide Powertrain Motor Design and Simulation
Johannes Sæbø Powertrain Motor Production
Kristian Ringheim Moe Powertrain Motor Thermal Management
Katla Maria Gudmundsdottı́r Embedded Electronics Group leader Embedded Electronics
Fartein Sveindal Embedded Electronics Vehicle Control Unit/ Propulsion Control Unit
Oscar Stentun Stadskleiv Embedded Electronics Battery Management System
Jarl Magnus Sæbø Embedded Electronics Battery Management System
Andreas Dahl Embedded Electronics Telemetry
Marvin Ademi Embedded Electronics Telemetry
Martin Fuglum Embedded Electronics Sensors
Seyed Husseini Embedded Electronics Sensors
Dibakar Sarker Structural Group Leader Structural
Jens Emil Valvik Structural Chassis
Julian Wiggers Structural Chassis
Magnus Nørstenes Structural Brakes
Filip Grønhaug Reierth Structural Brakes
Aleksander Auflem Structural Suspension
Tharanath Thiruthiyappan Structural Suspension
Hedda Gunnarsson Øyan Concept Group leader Concept
Gisela Cerron Concept Station Design
Mathea Skaar Concept Station Design
Bhuwan Panday Concept Route and City Planning
Sigurd Fagerholt Concept Route and City Planning
Vedran Simic Concept Pod Design
Francisco Soto Concept Pod Design
Johannes Gjerdåker Casing and Housing Group leader Casing and Housing
Frederico Reis Casing and Housing Aeroshell Design and Simulation
Isak Knutsen Casing and Housing Aeroshell Production
Aksel Moen Norberg Casing and Housing Aeroshell Design
Audun Sørheim Casing and Housing Aeroshell Simulation
Lars Halvor Hansen Casing and Housing Battery casing
Elias Öhman Casing and Housing Inverter casing
Olai Brevik Mykland Levitation R&D Group leader Levitation R&D / Simulation
Henrik Talstad Levitation R&D Levitation Sensors Engineer
Simen Moræus Wahlstrøm Levitation R&D Levitation Mechanical engineer
Diogo Viela Levitation R&D Levitation Mechanical engineer
Rémi Dujardin Software Group Leader of Software

15
Oskar Matre Software Backend
Simon Breivik Software Backend
Marcus Hagberg Software UAVCAN
Jørgen Jore Software Frontend
Eivind Njaastad Mentor Mechanical Systems
Magnus Johannessen Mentor Powertrain, Motor Design and Simulation
Ådne Børresen Mentor Marketing
Viktoria Shulga Mentor Marketing
Bendik Nyhavn Mentor Electrical Systems
Magnus Kolnes Oddstøl Mentor Electrical Systems

Table 2: Team members, Shift Hyperloop team 2022

16
B Appendix 2

Figure 6: Map of Hardangervidda, which has to be avoided, between Oslo and Bergen.
(Hardangervidda.com: https://hardangervidda.com/kart/)

17
C Appendix 3

Both the python script used in route planning and the routes plotted are found here:
https://drive.google.com/drive/u/3/folders/1dor2iQ48hjDC4oqXjOUG q2TNuNhieCS

18