Launch Vehicle Selection with TOPSIS: Multi-

Criteria Decision Making for Space Mission

Mücahid Alp Mert Cihan Tohum Doğukan Omaygenç
Turkish Air Force Academy Turkish Air Force Academy Turkish Air Force Academy
National Defence University National Defence University National Defence University
Istanbul, Türkiye Istanbul, Türkiye Istanbul, Türkiye
[email protected] [email protected] [email protected]

Hande Bodur Ersin Makas

2023 10th International Conference on Recent Advances in Air and Space Technologies (RAST) | 979-8-3503-2302-3/23/$31.00 ©2023 IEEE | DOI: 10.1109/RAST57548.2023.10197916

Turkish Air Force Academy Turkish Air Force Academy

National Defence University National Defence University
Istanbul, Türkiye Istanbul, Türkiye
[email protected] [email protected]

Abstract— From interplanetary to sending a basic spacecraft II. LAUNCH VEHICLE SELECTION
to space, the first goal is getting to space. Launch vehicle
selection is a paramount step of a space mission that many Equations Launch systems take the spacecraft from Earth
criteria such as launching, reliability of the launch system, to the desired orbit. They are two types which are disposable
environmental and meteorological conditions, availability, and reusable. Disposable ones are more frequently used, and
launch system and spacecraft compatibility, cost, lifting ability, they can be expanded, that is, add additional stages, and are
durability, relations between countries, and some other criteria for single-use. The other systems such as Space Shuttle or
are considered. Therefore, the launch system selection can be Falcon 9X, developed by Space X can be used again [2]. In
named a “multi-criteria decision-making problem” and it is an the selection process of launch systems for small satellites; the
evaluation of multiple alternatives and their characteristics. In first question is whether to launch a single load or a group
this study; TOPSIS (Technique for Order Preference by launch. Cost minimization is generally considered first for
Similarity to Ideal Solution) which determines the criteria small satellites; the trend is group launch and expandable
closest to the ideal solution together with AHP (Analytical launch systems are preferred [3]. Small satellites cannot
Hierarchy Procedure) to decide and choose the best alternative determine the launch schedule by themselves based on small
was studied for launch vehicle selection. satellite mission experience. Because they are low-cost
missions and are often planned as secondary payloads. The
Keywords— launch vehicle selection, TOPSIS, AHP-TOPSIS
launch program is usually determined by an expensive and
I. INTRODUCTION large primary payload. Therefore, undesired delays may occur
in the launch program due to the primary payload. This delay
From sending satellites to sending people to space, from may cause to increase in the cost of the mission. For example,
studying other planets to interplanetary travel, the first goal of in 2000, the Canadian Space Agency had to look for
all space missions is to reach space safely. In this study; within alternative launch options due to the ongoing delays for the
the scope of space mission architecture, the launch system RADSAT-2 mission. Fortunately, there are more launch
selection was studied. In a new space mission system analysis options today. Nevertheless, undesired delays should be taken
and mission design, assessment of launch vehicle options is into account when deciding to launch multiple or single ones
usually done quickly. Such calculations allow engineers to in the launch system selection [4].
evaluate the launch system for orbit performance, cost, and
weight during the design and development phase of the task. The launch process can greatly restrict spacecraft design.
It is usually made with tables and reference manuals. However, The main restrictions are the lifting capacity of the launch
this method can be time-consuming and limited [1]. In space vehicle and the environment in which the satellite is exposed
missions, engineers do careful calculations and evaluations during launch. A launch system consists of a single or multi-
before and after the launch system selection. They also staged basic launch vehicle and a ground support
consider possible launch and program changes of launch infrastructure facility. It changes its speed to place the launch
vehicles and spacecraft. The correct analysis and selection of vehicle in its orbit and performs its rise in atmospheric density,
launch options for the entire space mission, which is a protecting the spacecraft from external forces. Launch
challenging and costly step of a mission directly affects the vehicles and spacecraft are managed as separate organizations
task schedule and consequently its success. In this paper, the in the space industry, and launch vehicle and spacecraft
launch system selection stages, criteria, and how they are interfaces must be compatible. The durability and cost of the
evaluated are examined. In addition, which criteria to be launch system for the success of a mission must be understood
considered in the choice of launch system for a space mission by the planners [5].
were studied. This selection process has multiple criteria and
The first step of the launch system selection process is to
the priorities of the criteria which means a multi criteria
determine the requirements and goals of the mission because
decision making process. TOPSIS’s innovative and dynamic
the requirements and objectives of the mission will determine
decision making method and its steps will be explained and
the family, route, and performance of the launch vehicle to be
implemented in the selection of launch vehicles.
carried out from the appropriate launch units. The
determination of the task also changes the terms to be used in
this regard. Understanding the real mission requirement is

XXX-X-XXXX-XXXX-X/XX/$XX.00 ©20XX IEEE

Authorized licensed use limited to: UNIVERSITY PUTRA MALAYSIA. Downloaded on May 27,2024 at 16:44:05 UTC from IEEE Xplore. Restrictions apply.
crucial as it will dictate the launch strategy. Another critical equator, the more it is close to the equator, the more
issue is whether the spacecraft to be launched will be launched achievable for the desired orbit inclination [9].
only by itself or whether it will share the same launch vehicle
with other secondary cargoes. It may be costly for the Black and Nurre stated that the choice of launch system is
spacecraft to have only its own launch system, but in this case, critical in their duties and costs. According to NASA's small
the possibility of problems with other secondary cargo satellite program, launch costs account for 21% of the total
spacecraft will affect the launch. The confidentiality of the satellite program. According to their studies, the most
load can also create a special launch need. On the other hand, important criterion in the selection of the launch system is cost.
shared launches reduce the cost per spacecraft. The interaction Therefore, the optimal choice in the launch system selection
of the loads with each other is important in shared launches. If will reduce the launch cost. The main criteria in the selection
we place them in series, for example, if more than one of the launch system are weight, altitude, orbital inclination,
spacecraft is launched, it can be considered if the target orbital the number of satellites to be placed in each orbital plane, and
altitudes and gradients are compatible [5]. The trend in small the availability and cost of the launch system [10]. In the
satellites is group launch, as cost minimization is often taken space-sizing study, it was stated that the selection of the
into account and can be expanded accordingly. Launcher launching system was limited in length-setting and wet mass
determination. Also, the launch cost for a medium-sized
systems that can be added to the upper stage are preferred.
After the task requirement is determined, the job requests for commercial spacecraft may be more than 50% of the total cost
this need are determined. For LEO tasks, these usually include [11].
orbital parameters such as orbital altitude, tilt angle, and The launch window is determined according to when we
ascending node. In addition, the estimated load weight and will launch the spacecraft directly from a particular launch
dimensions are decisive for the launch system. With the unit to a desired orbit. Orbital planes are complicated and
mission concept, these parameters are detailed as the number require more fuel. In addition to the launch time, the physical
of spacecraft to be sent, life expectancy, change strategy and condition of the desired orbit and launch unit is also evaluated.
management, and data exchange. The desired launch date can The launch window determination is briefly explained as
also be a selection parameter. Because the suitability of the follows. First, the orbital plane and the launch location must
launch vehicle and launch facility also affects the launch be determined. The meaning of the launch window is to
schedule [5]. Launch system selection criteria for space choose a specific solar time. Task planners refer to Greenwich
missions performed in the Stanford University spacecraft / England as the international clock time when determining
design class in 2007 were determined as low cost, lifting launch time [12]. Again, in a study for small satellite launch
capability, and maturity. In the task, in which seven different options, alternatives for the launch system selection were
launch systems were evaluated, the selection process of the evaluated with the Analytic Hierarchy Processing approach
constantly repeating launch system was handled. Thus, the using Expert Choice software [13].
design team constantly worked to adapt their subsystems to
the launch system [6]. In his study, Greenberg examined the In the Lafleur study, he applied the probabilistic AHP and
DAMSAT model, which provides financial evaluation for TOPSIS method for multi-attribute decision-making under
communication satellites. In the model comparing launch uncertainty to select the launch system that would take the
systems, the cost, insurance, reliability, availability, and risk small military satellite mission into circular orbit [14]. In
factors of launch vehicles that will bring communication another example, Tetik and Das applied multi-criteria
satellites into geostationary orbit were compared and analyzed. decision-making methods for the selection of the launch
Actions and policies of governments, technological system and evaluated the selection of the launcher with AHP
investments, and even the launching priorities of commercial and PROMETHEE [15]. In this study, we examined a multi-
launches from governmental facilities can affect the decision criteria decision making problem which is launch system
maker's final decision on the launch system. It is stated that selection.
many effects, from government effects to the reliability of the In this paper, AHP and TOPSIS are studied on a space
launch system, should be evaluated before investing in an issue. AHP is selected to define the weights of the criteria.
expensive space transportation service that will bring the Though the multi-criteria decision making method applied in
communication satellite to the desired orbit [7]. The suitability this paper is a hybrid (AHP and TOPSIS), we have just used
of a launch system depends on its reliability, production the weight results of AHP, which is widely used for it, and
capacity, the ability of launch operations systems to support then TOPSIS steps have been conducted. Apart from the wide
the targeted launch frequency, current launch commitments, acceptance of the methods, they are simple enough to explain
and its proven waiting time after a failure [5]. In some a space subject as multi-criteria decision making. Our goal is
missions, security and confidentiality may also be a priority, to show that it is possible to solve a fundamental and important
and therefore non-commercial launch systems can be step of a space mission analysis and design process with multi-
evaluated. Three main criteria were evaluated in the selection criteria decision making. Before starting the selection process
of the launch system for the SandiaSat mission. The first is steps, we are going to explain the methods of TOPSIS
cost, the second is security, and the third is launch availability. (Technique for Order Preference by Similarity to Ideal
While evaluating launch systems, allowing sufficient mass Solution), which determines the criteria closest to the ideal
margin to get the payload mass into the required orbit was also solution, and the AHP (Analytic Hierarchy Process) methods.
considered an important factor [8].
A. TOPSIS
The latitude of the launch site affects the orbit parameters.
TOPSIS (Technique for Order Preference by Similarity to
In general, launches from the southern part of the globe are
Ideal Solution) is a multi-criteria decision making method. For
made for low-polar orbits, and eastern launches have
two ideal solutions in the method; “The positive ideal solution
advantages from the Earth cycle. Launching from near the
minimizes the cost while maximizing the benefit, while the
negative ideal solution also minimizes the cost” [16] Finding

Authorized licensed use limited to: UNIVERSITY PUTRA MALAYSIA. Downloaded on May 27,2024 at 16:44:05 UTC from IEEE Xplore. Restrictions apply.
only the closest criterion to the positive ideal condition is not alternatives, for the problem, in the bottom step, there are
enough to satisfy the decision maker and if practically possible, criteria that are used to evaluate the alternatives. The number
the alternative the most beneficial and the least risky is of criteria for obtaining pairwise comparisons should be
preferred. Critical values and importance weights in the determined correctly and each criterion should be defined
TOPSIS process are numerical values [17]. correctly.
Step 1, Normalization of the Decision Matrix: The
normalization vector, shown as equation (1), specified in the In the first stage, the subject and purpose of decision-
normalization formula is used for the normalization of the making is the hierarchical staging of the relevant decision
decision matrix. elements. Decision indicators and decision choices are the
𝑥𝑖𝑗 elements of decision-making.
𝑟𝑖𝑗 = 𝑖 = 1, . . . , 𝑚𝑗 = 1, . . . , 𝑛 (1)
√∑𝑚
𝑖=1
2
𝑥𝑖𝑗
In the second step and to conduct a paired comparison, a
Step 2, Weights of Decision Matrix: The criterion value of an questionnaire should be designed and distributed among the
alternative is multiplied by the weights of the criteria. It is
respondents (experts, users, etc.) to collect their opinion.
equation (2).
Notably, each decision maker entered their desired amount for
𝑣𝑖𝑗 = 𝑤𝑖𝑗 . 𝑟𝑖𝑗 𝑖 = 1, . . . , 𝑚𝑗 = 1, . . . , 𝑛 (2) each member and then individual judgments (for each one of
the pair comparisons) using their geometrical average. The
Step 3, Determining the Positive and Negative Ideal Solution: scale ranges from one to nine; here it is implied that the two
𝐴+ and 𝐴− selected from members of normalized matrix
elements are equally important. On the other side, the number
weights. While 𝐽1 the benefit criterion is 𝐽2 the cost criterion,
nine shows that one principle is extremely important than the
the determination of positive and negative ideal solutions is as other one in a binary matrix. The data analysis procedure
following equation (3). includes the following steps. First, the binary comparison
  matrix which is named matrix A is extracted from the data
v , v ,..., v ,..., v  =  m a x v | j  J  ,  min v | j  J |i 1,..., m
*
1
*
2
*
j
*
n ij 1 ij 2
collected from the interviews. The basic right eigenvector of
 i  i   matrix A is calculated as ‘w’. If 𝑎𝑖𝑗 . 𝑎𝑘𝑗 = 𝑎𝑖𝑗 was not verified
𝐴− = {𝑣1− , 𝑣2− , . . . , 𝑣𝑗− , . . . , 𝑣𝑛− } (3) for all. k, j, and i the Eigenvector method is selected. If the
matrix is inconsequent and in case of missing consistency, the
    
=  min vij | j  J1  ,  max vij | j  J 2 |i 1,..., m  double comparisons matrix can not be used deradicalization
 i   i   to get 𝑊𝑖 . For a positive and inverted matrix, the Eigenvector
Step 4, Separation Measurements: Calculation of distances technique contained in can be used and shown as equation 7.
between negative and positive ideal solutions which are
eT = (1,1,…,1) (7)
shown as equations (4) and (5). 𝐴𝑘 .𝑒
2 𝑊=
𝑆𝑖+ = √∑𝑛𝑗=1 (𝑣𝑖𝑗 − 𝑣𝑗∗ ) 𝑖 = 1, . . . , 𝑚 (4) 𝑒 𝑇 .𝐴𝑘 .𝑒
To reach a convergence among the set of answers in the
Distance from a negative ideal solution, successive iteration of this process, the computation must be
𝑆𝑖− = √∑𝑛𝑗=1 (𝑣𝑖𝑗 − 𝑣𝑗− )
2
𝑖 = 1, . . . , 𝑚 (5) repeated several times to decide when an incompatible matrix
is encountered. Next, the following formula is applied to
convert the raw data to mean absolute values and the
Step 5, Calculation of Closeness to Ideal Solution: 𝐶𝑗 0,1 shows the
normalized weight W = ( 𝑊1 , 𝑊2 , 𝑊3 … 𝑊𝑛 ) and shown in
degree of proximity. The equation is (6),
𝑆− equation 8.
𝐶𝑖∗ = ∗ 𝑖 − 𝑖 = 1, . . . , 𝑚 (6) Aw = λmax w, λmax ≥ n
𝑆𝑖 +𝑆𝑖
∑𝑎𝑗 𝑤𝑗 −𝑛
𝜆𝑚𝑎𝑥 = (8)
Step 6 Determining the Order of Preference: Alternatives are 𝑛
prioritized according to the value of 𝐶𝑖∗ or the highest value of 𝐶𝑖∗ is A={𝑎𝑖𝑗 } with 𝑎𝑖𝑗 =1/ 𝑎𝑖𝑗
chosen [18]. A: binary comparison
w: normalized weight vector
𝜆𝑚𝑎𝑥 : maximum eigenvalue of matrix A
B. AHP
𝑎𝑖𝑗 : numerical comparison between the values i and j

AHP (Analytical Hierarchy Procedure) is a multi-criteria In the next step, to verify the results of the AHP, the
decision making method [19]. This method provides the consistency ratio (CR) is calculated using the formula CR =
solution to complex problems by taking into account the CI / RI, where the consistency index (CI) is measured by the
mathematics, quantitative and qualitative criteria, experience, following equation 9.
knowledge, and intuition of the decision maker. In this way, 𝜆 −1
the decision maker can evaluate different alternatives and 𝐶𝐼 = 𝑚𝑎𝑥 (9)
𝑛−1
make an original choice. The method, which was first
introduced by Myers and Alpet in 1968, was later developed The value of RI is related to the size of the matrix. It
by Saaty in 1977 [20]. should be noted that a consistency ratio of less than 0.10
confirms that the comparison results are acceptable.
The hierarchy in the AHP stages consists of three levels. Stakeholders compare the importance of the criteria, two at a
At the top of the hierarchy is the ultimate goal, in the middle, time, through bilateral comparisons. For example, do you care
there are all of the possible solutions, which are called more about work benefits or a short commute, and how much

Authorized licensed use limited to: UNIVERSITY PUTRA MALAYSIA. Downloaded on May 27,2024 at 16:44:05 UTC from IEEE Xplore. Restrictions apply.
more? AHP converts these evaluations into numbers that can A. AHP Analysis for Weights
be compared with all criteria. This quantification ability
distinguishes AHP from other decision-making techniques. At The weights of the criteria in the study were calculated by
the last stage of the process, numerical precedence is making a double comparison from the steps of the Analytical
computed for each of the alternative options. These numbers Hierarchy Process (AHP) method. The binary comparison
represent the most desired solutions according to the values of survey was applied to experts and is shown in Table 1 for if
all users. reliability is more important and Table 2 shows the result if
cost is the more important criterion.
III. LAUNCH VEHICLE SELECTION WITH TOPSIS
TABLE I. LAUNCH SYSTEM SELECTION CRITERIA DOUBLE
Decision-maker preferences and trends in launch system COMPARISON 1
selection were examined. We classified the decision-maker kg $ R Lat m3
into three choices. The first is classified as those who think the Weight (kg) 1,00 0,33 0,20 3,00 5,00
cost is minimal, the second is those who think the risk is Cost ($ 1000) 3,00 1,00 0,33 5,00 5,00
minimal, and the third is the ones who consider the balance of Reliability (R) 5,00 3,00 1,00 5,00 7,00
cost and risk. It is understood that in general, decision-makers Geographical Suitability 0,33 0,20 0,20 1,00 3,00
focus on these two issues. Those who consider the minimum (Latitude)
cost are primarily focused on cost and do not want to pay high Eligibility For Payload (m3) 0,20 0,20 0,14 0,33 1,00
costs for low risk. In turn, decision-makers who do not want
to take risks can ignore how expensive the launch system is TABLE II. LAUNCH SYSTEM SELECTION CRITERIA DOUBLE
COMPARISON 2
for the reliability of the launch system.
kg $ R Lat m3
After searching and mentioning criteria for launch vehicle Weight (kg) 1,00 0,33 0,20 3,00 5,00
selection, the criteria that have been studied in this paper are Cost ($ 1000) 3,00 1,00 0,33 7,00 9,00
as follows. Launch vehicles and criteria can be extended and Reliability (R) 5,00 3,00 1,00 3,00 7,00
changed. The number of criteria has been taken in a reasonable
Geographical Suitability 0,33 0,20 0,20 1,00 3,00
minimum number to make simple calculations and it helps to (Latitude)
make the subject clear. However launch vehicle type and Eligibility For Payload 0,20 0,20 0,14 0,33 1,00
company can surely differ in mission needs, the main idea (m3)
never changes but the values of the launch systems.
No matter how consistent systematic AHP has, the
𝑔1 ; The Load That The Launch System Can Carry: The realism of the results will naturally depend on the consistency
spacecraft we want to launch can be the main load or the of decision-makers between the criteria. Therefore, the
secondary load. Only the desired single payload can be consistency analysis rate (CR) is calculated by making a
launched or it can use the launch vehicle with secondary consistency analysis. This measure not only detects decision-
payloads. As a result, it is a requirement that the launch unit maker’s misinterpretations in their binary comparisons and
lifts the desired weight. allows them to reduce careless mistakes but also displays
exaggerated evaluations. The acceptable upper limit for the
𝑔2 ; Launch System Cost: Launch system cost should be consistency rate is 0.10. This rate indicates that the probability
of comparing the criteria inconsistently, completely, and
minimum. Among the costs allocated for the task, it should be
randomly is 10%. In other words, it means that the decision-
at the values that can meet the costs considered for the launch
maker makes binary comparisons consistent. If the
operation. In this study, kg. cost-per-use features were used.
consistency rate is greater than 0.10, the decision maker is
recommended to review their comparisons. The essence of the
𝑔3 ; Reliability: The launch system should be reliable,
CR calculation is based on comparing the number of criteria
including country, rocket, unit, and firm. It is necessary to (n) and a coefficient called the base value (λ). It is calculated
evaluate how long it has been launched before, and how as λ = 5.3502, CI = 0.08799, and CR = 0.0782 and is consistent
successful it has been, how long it will be able to fix and because it is less than 10%. After binary compression and
launch the problem after the problems that occur or may occur. implementation for Table 1 and Table 2, the results of the
weights are shown in Table 3 and Table 4 respectively.
𝑔4 ; Geographical Conformity: Since the place to be
launched will affect the orbital parameters, the appropriate
geographic launch unit must be selected to conduct the desired
task.

𝑔5 ; Suitability for the Payload: It is important whether the

launch system is suitable for the payload. The adapter, the
physical and electrical interface of the launch system, the
space and payload allocated for the payload, and protective
properties from the launch to the space environment should be
appropriate.

Authorized licensed use limited to: UNIVERSITY PUTRA MALAYSIA. Downloaded on May 27,2024 at 16:44:05 UTC from IEEE Xplore. Restrictions apply.
 TABLE III. LAUNCH SYSTEM SELECTION CRITERIA WEIGHTS 1 minimum for the launch system cost, the maximum for
W reliability, the maximum for geographical suitability, and the
Weight (kg) 0,146 maximum for suitability for the payload and the result of this
Cost ($ 1000) 0,258 step are shown in Table 8.
Reliability (R) 0,475
Geographical Suitability (Latitude) 0,079
Eligibility For Payload (m3) 0,042

TABLE IV. LAUNCH SYSTEM SELECTION CRITERIA WEIGHTS 2

W
Weight (kg) 0,146
Cost ($ 1000) 0,466
Reliability (R) 0,28
Geographical Suitability (Latitude) 0,074
Eligibility For Payload (m3) 0,035

B. TOPSIS Steps on Launch Vehicle Selection

Satellites sent to LEO orbital missions for various

purposes have been examined. Selected launch systems in
this study have enough weight lift capability. Payload
environment is taken as the volume capability of the vehicle
and latitude has been considered as the distance from the
equator which means the minimum value is desired. For
reliability, the desired value will be a maximum of 1, and sure
the cost is minimum. The decision matrix created with the
values taken from the reference sources for the selected
alternatives is shown in Table 5. Alternative platforms in this
study are Atlas V, Falcon 9, Soyuz, and Ariane 5.

TABLE V. DECISION MATRIX

LV Weight Cost Reliabilit Latitude Payloa
(kg) ($1000) y (R) d (m3)
Atlas V 18850 13 0,99 28,3 183,45
6
Falcon 9 22800 3 0,91 28,3 95
Soyuz 24400 5,2 0,98 62,48 243,67
Ariane 5 18000 7,2 0,89 5,32 243

The sum of the squares of the values belonging to each

criterion of the decision matrix (column values of taking the
square root of the sum of squares), dividing the corresponding
element of the column by this output value the standard
decision matrix is obtained with. The second step of the
TOPSIS method was created by using the values in Table 5
and the results are shown in Table 6. By using the normalized
matrix, positive ideal, and negative ideal solution sets are
obtained for each criterion according to the evaluation purpose
of the criterion of interest. In terms of utility, the evaluation
criteria are the largest values of the columns of the positive
ideal solution decision matrix and the smallest values of the
columns of the negative ideal solution decision matrix. In
terms of cost, the evaluation criteria are the smallest values of
the columns of the positive ideal solution decision. The results
obtained when the third step of the TOPSIS method is applied
are shown in Table 7. Positive and negative ideal solutions are
found in the 4th step of TOPSIS. Accordingly, the desired
positive ideal solutions; It has been determined as the
maximum for the load that the launch system can carry, the

Authorized licensed use limited to: UNIVERSITY PUTRA MALAYSIA. Downloaded on May 27,2024 at 16:44:05 UTC from IEEE Xplore. Restrictions apply.
 TABLE VI. TOPSIS STEP 2 NORMALIZED MATRIX
Launch Vehicle Weight (kg) Cost ($ 1000) Reliability (R) Latitude Payload (m3)
Atlas V 0,444976321 0,811107106 0,524647373 0,380428828 0,45706417

Falcon 9 0,538220695 0,187178563 0,482251625 0,380428828 0,23668398

Soyuz 0,575990569 0,324442842 0,519347904 0,839900818 0,60708195

Ariane 5 0,424911075 0,449228551 0,471652689 0,071515243 0,605412705

TABLE VII. TOPSIS STEP 3 DISTANCES TO NEGATIVE IDEAL SOLUTION

0,146 0,258 0,475 0,079 0,042
Launch Vehicle Weight (kg) Cost ($ 1000) Reliability (R) Latitude Payload (m3)
Atlas V 0,064966543 0,209265633 0.249207502 0.030053877 0.019196695

Falcon 9 0,078580222 0,048292069 0.229069522 0.030053877 0.009940727

Soyuz 0,084094623 0.083706253 0.246690255 0.066352165 0.025497442

Ariane 5 0,062037017 0.115900966 0.224035027 0.005649704 0.025427334

TABLE VIII. TOPSIS STEP 4 POSITIVE AND NEGATIVE IDEAL SOLUTION

Launch Vehicle Weight (kg) Cost ($ 1000) Reliability (R) Latitude Payload (m3)
Positive Ideal 0,084094623 0.048292069 0.224035027 0.005649704 0.009940727
Solution

Negative Ideal 0,062037 0.209265633 0.249207502 0.066352165 0.025497442

Solution

Authorized licensed use limited to: UNIVERSITY PUTRA MALAYSIA. Downloaded on May 27,2024 at 16:44:05 UTC from IEEE Xplore. Restrictions apply.
 In step 5 of TOPSIS, the difference to the ideal solution design, manufacture and testing”, Springer Science & Business
Media, 2012.
is calculated by applying the square root (Euclidean) of the
[4] Grocott, Simon, Robert Zee, and Jaymie Matthews. "Exploring the
difference between the weighted normalized values in step Mysteries of the Cosmos on the MOST Microsatellite Mission."
3 of TOPSIS and the positive ideal solution and negative (2003).
ideal solutions in step 4. For TOPSIS, only the distance to [5] Larson, Wiley J., and James Richard Wertz, eds.; “Space mission
analysis and design”, Vol. 3. Torrance, CA: Microcosm, 1992.
the ideal solution points, and their differences are required,
[6] Barrientos, Francesca A., Irem Y. Tumer, and David G. Ullman.
and the distances are maximum and minimum values "Modeling uncertainty reduction in concurrent engineering design
according to the positive and negative ideal solution found teams." International Design Engineering Technical Conferences
for the criteria in the TOPSIS assessment. These values are and Computers and Information in Engineering Conference. Vol.
48051. 2007.
taken as intervals in the criteria. [7] Greenberg, Joel. "Selecting a launch vehicle for GEO missions."
16th International Communications Satellite Systems Conference.
The first result is if the reliability is the most important 1996.
and in accordance with this Soyuz has been selected with a [8] Whitmore, Stephen A., and Tyson K. Smith. "Launch and
slight difference from Falcon 9. In this case, the selection Deployment Analysis for a Small, Medium Earth Orbit,
Technology-Demonstration Satellite." Journal of Spacecraft and
sequence is as follows; Soyuz, Falcon 9, Ariane 5, and Rockets 46.2 (2009): 449-458.
Atlas V. [9] Brown, Charles D. Spacecraft mission design. AIAA, 1998.
[10] Colombi, John M., et al. "Optimal launch manifesting for
The second result is if the cost is the most important and heterogeneous disaggregated satellite constellations." Journal of
in accordance with this Falcon 9 has been selected. In this Spacecraft and Rockets 54.3 (2017): 582-591.
case, the selection sequence is as follows; Falcon 9, Soyuz, [11] Hassan, Rania, and William Crossley. "Spacecraft reliability-based
design optimization under uncertainty including discrete variables."
Ariane 5, Atlas V Journal of Spacecraft and Rockets 45.2 (2008): 394-405.
[12] Sellers, Jerry Jon, et al. Understanding space: An introduction to
IV. CONCLUSION astronautics. Primis, 2000.
[13] Matthew A. Bille Erika Lishock; “International Space Industry
There may be more than one criterion to be considered report (ISIR).
and evaluated. In this study, TOPSIS and AHP methods are [14] Lafleur, Jarret M. "Probabilistic AHP and TOPSIS for multi-
attribute decision-making under uncertainty." 2011 Aerospace
used to help decision makers for a space subject. Weight Conference. IEEE, 2011.
calculations were made with the help of binary comparison, [15] Tetik, Taha, and G. Sena Daş. "Launch vehicle selection for a
which is a step in the AHP method and is widely used. geostationary communication satellite using data envelopment
While making weight calculations of alternative criteria for analysis." 2017 8th International Conference on Recent Advances in
determining and evaluating our preferences, a survey was Space Technologies (RAST). Ieee, 2017.
applied to experts and academics. If the values are very [16] Hu, Jiantao, et al. "A modified weighted TOPSIS to identify
influential nodes in complex networks." Physica A: Statistical
high points in the survey applied, they were not taken into Mechanics and its Applications 444 (2016): 73-85.
account. [17] Y.J. Lai, T.Y. Liu ve C. L. Hwang; “TOPSIS For MCDM”,
European Journal of Opertional Resarch, Volume 76, Issue 3, 11
There are many studies in the literature conducted with August 1994, Pages 486-500.
these methods. Many decision making problems were tried [18] Okul, Deniz, Cevriye Gencer, and Emel Kizilkaya Aydogan. "A
method based on SMAA-topsis for stochastic multi-criteria decision
to be solved with these methods. However, in this study, making and a real-world application." International Journal of
the subject of space sciences is discussed. It is also Information Technology & Decision Making 13.05 (2014): 957-978.
emphasized that the task design, which is a space science [19] Saaty, Thomas L. "A scaling method for priorities in hierarchical
study and which concerns satellite technologies, is the stage structures." Journal of mathematical psychology 15.3 (1977): 234-
of the task concept, which has to be determined in advance, 281.
the cost and timeliness of the task, the choice of the launch [20] Wind, Yoram, and Thomas L. Saaty. "Marketing applications of the
analytic hierarchy process." Management science 26.7 (1980): 641-
system and the right decision to be made. In a costly study 658.
such as launching a satellite into space with very precise
calculations, the launch selection process and its features
are tried to be explained. The launch vehicle selection
process was taken as a multi-criteria decision making
problem and tried to be solved with a known multi-criteria
decision-making method.

REFERENCES

[1] Lafleur, Jarret M., Elizabeth S. Fleming, and Joseph H. Saleh.

"Response surface equations for expendable launch vehicle payload
mass capability." Journal of Spacecraft and Rockets 49.1 (2012):
185-189.
[2] Maini, Anil K., and Varsha Agrawal;” Satellite technology:
principles and applications”, John Wiley & Sons, 2011.
[3] Abdelal, Gasser F., Nader Abuelfoutouh, and Ahmed H Gad. ;”
Finite element analysis for satellite structures: applications to their

Authorized licensed use limited to: UNIVERSITY PUTRA MALAYSIA. Downloaded on May 27,2024 at 16:44:05 UTC from IEEE Xplore. Restrictions apply.