Deriving Analytical Solutions Using Symbolic Matrix Structural Analysis:

Part 1 – Continuous Beams

Vagelis Plevris1* and Afaq Ahmad2

1
Department of Civil and Environmental Engineering, Qatar University
P.O. Box: 2713, Doha, Qatar
e-mail: [email protected]
2
Department of Built Environment, Oslo Metropolitan University
P.O. Box 4, St. Olavs Plass, NO-0130, Oslo, Norway
e-mail: [email protected]

*Corresponding author

Keywords: Matrix structural analysis, symbolic MSA, continuous beams, structural analysis,
sensitivity analysis, influence lines, analytical solutions, closed-form solutions.

Abstract. This study investigates the use of symbolic computation in Matrix Structural
Analysis (MSA) for continuous beams, leveraging the MATLAB Symbolic Math Toolbox. By
employing symbolic MSA, analytical expressions for displacements, support reactions, and
internal forces are derived, offering deeper insights into structural behavior. This approach
facilitates efficient and scalable sensitivity analysis, where partial derivatives of outputs
concerning input parameters can be directly computed, enhancing design exploration. The
development includes an open-source MATLAB program, hosted on GitHub, enabling
symbolic analysis of continuous beams subjected to point and uniform loads. This approach is
invaluable for both engineering practice and pedagogy, enriching the understanding of
structural mechanics and aiding in education by illustrating clear parameter relationships. The
program supports deriving influence lines and identifying maximum response values.

1 Introduction to Symbolic Matrix Structural Analysis for Structural Engineering

The Finite Element Method (FEM), which recently marked its 80th anniversary since its
inception [1], has been a foundational tool in structural analysis, offering a robust numerical
framework for solving complex engineering problems [2]. Traditional FEM approaches are
predominantly numerical, relying on computations to approximate structural behavior under
specified loads and boundary conditions. While effective for producing detailed, problem-
specific results, these methods have limitations, particularly in flexibility and generality.
Numerical FEM solutions are generally applicable to a specific set of inputs, such as material
properties, geometry, and external loads [3]. Modifications to these inputs necessitate a
complete re-computation of the system, which can be both time-consuming and
computationally intensive, especially for large-scale structures. Additionally, numerical results
often obscure the relationships between key parameters, making it difficult to develop a deeper
understanding of structural behavior.

1
 Deriving Analytical Solutions Using Symbolic Matrix Structural Analysis: Part 1 – Continuous Beams

FEM applied to linear-type structures, such as beams, trusses, and frames, is commonly
referred to as Matrix Structural Analysis (MSA). Both methods share the same underlying
principles and origins. In MSA, analytical expressions for the stiffness matrix of each element
can often be derived directly, bypassing the need for numerical integration techniques such as
Gaussian quadrature. This feature enhances the process by making the derivation simpler and
more insightful. One of the primary drawbacks of traditional numerical FEM and MSA is their
reliance on predefined boundary conditions and loading scenarios. For every new condition,
the entire model must be regenerated and recalculated, leading to a repetitive and time-
consuming workflow. This limitation hinders the method’s adaptability and restricts its
application in tasks requiring real-time analysis or optimization [4], where multiple
configurations need to be evaluated quickly and efficiently [5].
Symbolic computation offers a promising alternative to these challenges. Unlike numerical
computation, which yields specific numerical values, symbolic computation allows for the
manipulation and solution of mathematical expressions in their exact, algebraic form. This
approach opens new avenues for deriving analytical solutions and gaining deeper insights into
structural behavior.
MATLAB is extensively utilized in structural engineering due to its robust numerical
algorithms, which enable the resolution of a variety of engineering challenges, including matrix
analysis of structures [6-8], structural dynamics [9-11], structural optimization [12-14], and
more [15]. While MATLAB is primarily known for its numerical computing capabilities, it also
includes features for symbolic computation. The MATLAB Symbolic Math Toolbox [16]
significantly enhances MATLAB’s functionality, allowing it to handle symbolic computation
[17, 18]. The toolbox supports a wide range of symbolic operations, including algebraic
simplifications, differentiation, integration, equation solving, and matrix manipulation [19]. It
is particularly valuable in engineering, physics, and mathematics, where it facilitates the
analysis of complex systems by providing exact, parameterized solutions that can be easily
manipulated and interpreted. For researchers and educators, the Symbolic Math Toolbox is an
indispensable resource, as it enhances the ability to explore theoretical concepts, derive closed-
form solutions, and present results in a more intuitive and generalizable manner. Its integration
with the program’s numerical environment also allows for seamless transitions between
symbolic and numerical analysis, offering a comprehensive platform for tackling both
theoretical and practical problems.
While other software and programming languages like Mathematica [20], Maple [21], and
SymPy (Python) [22] also offer symbolic computation capabilities, this study focuses on the
MATLAB Symbolic Math Toolbox [16, 23], which was used to develop the symbolic analysis
open-source code of this work. However, similar programs can be developed using these
alternative platforms, following the same underlying principles. In structural analysis, the
toolbox facilitates the expression of MSA solutions in symbolic form, where the system’s
response is represented by algebraic expressions involving key parameters related to material
or section properties. This method is beneficial for analyzing structural elements such as beams,
trusses, and frames, where the stiffness matrix for each element can be derived symbolically,
eliminating the need for complex numerical integration. By combining the symbolic stiffness
matrices of individual elements and expressing the force vector symbolically, fully symbolic

2
 Vagelis Plevris and Afaq Ahmad

solutions for displacements, element internal forces, or support reactions can be obtained. This
enables flexible manipulation of solutions to accommodate various loading conditions,
boundary constraints, or material properties without the need for repeated recalculations.

1.1 Literature Review

There have been relatively few attempts in the literature to address stiffness matrices
symbolically. Korncoff and Fenves [24] made an early effort to develop a symbolic processor
aimed at assisting in the generation of stiffness matrices for finite elements, despite the limited
computational resources available at the time. Their results, however, highlighted several
promising avenues for future research, both within the specific domain of finite element
analysis (FEA) and in the broader application of symbolic processing techniques. Leff and Yun
[25] presented a system for generating global stiffness matrices where elements are expressed
as functions of shape parameters. Their approach builds on STRUDL syntax, but with a
significant difference: instead of fixed values, joint coordinates, material properties, and forces
are entered as parameterized expressions. The resulting stiffness matrix retains these parameter
dependencies, providing flexibility in the analysis. This method allows engineers to explore
how changes in structural parameters affect the stiffness matrix, offering a more versatile and
adaptive framework for FEA, particularly beneficial for sensitivity studies and parametric
design exploration.
Nagabhushanam et al. [26] developed a specialized symbolic manipulation package in
FORTRAN for generating elemental matrices in FEA. The package allows users to perform
symbolic manipulations through simple, user-friendly commands. With a modular structure and
minimal memory requirements, it efficiently handles large-order matrices, even on personal
computers. The package supports various element geometries and shape functions, including
isoparametric elements, and offers a multilevel operator facility to perform several
manipulations with a single command. This compact tool has been successfully applied to
generate elemental stiffness, flexibility, and nodal force matrices, demonstrating its utility in
symbolic FEA. Tummarakota and Lieh [27] addressed the need for efficient algorithms to
model and predict the behavior of multibody structural systems in applications such as
aerospace, robotics, and automotive engineering. They utilized a computer-aided symbolic
method to formulate equations of motion based on Lagrange’s method, offering greater physical
insight compared to traditional numerical approaches. The generated equations are
automatically converted into FORTRAN code, enabling simulations and control synthesis. The
study demonstrated the effectiveness of this approach through two examples: a slider-crank
mechanism and an aircraft model, which are solved using the Runge-Kutta-Fehlberg method
for numerical integration.
Eriksson and Pacoste [28] discuss the use of symbolic software in developing finite element
procedures, particularly for complex problems involving higher-order instabilities requiring
precise formulations. The authors highlight that symbolic tools enhance the efficiency and
clarity of procedural development, enabling effective comparisons between various element
assumptions. The research includes beam formulations for plane and space models, allowing
analytical verification of equivalence between displacement and co-rotational contexts.
Symbolic derivation also simplifies handling finite space rotations and supports the systematic
derivation of local displacements from global variables. Amberg et al. [29] describe the

3
 Deriving Analytical Solutions Using Symbolic Matrix Structural Analysis: Part 1 – Continuous Beams

development of a toolbox in Maple for generating finite element codes from symbolic
mathematical specifications, facilitating 1D, 2D, and 3D simulations. This toolbox has
significantly accelerated research in areas such as thermocapillary convection, welding, and
crystal growth by reducing the time from conceptualization to a functioning simulation to mere
hours. The approach promotes flexibility, transparency, and thorough documentation, enabling
researchers to easily modify models and focus on physical insights and numerical properties
while minimizing errors and debugging.
Pavlovic [30] discussed the use of symbolic computation in structural engineering,
highlighting its emergence as a powerful tool alongside traditional numerical methods. He
reviewed past applications where symbolic computations have been underutilized and
emphasized the potential for significant advancements in areas of classical structural analysis.
The author argues that symbolic computation invigorates classical analytical techniques,
offering new opportunities to address complex structural problems. He proposes a more
balanced approach that integrates both symbolic and numerical methods, demonstrating their
complementary strengths in solving structural mechanics problems efficiently and accurately.
Skrinar and Pliberšek [31] derived a symbolic stiffness matrix and load vector for a slender
beam with multiple transverse cracks under uniform loading. Using the principle of virtual
work, they provided closed-form analytical expressions that facilitate fast and straightforward
evaluations. This approach, which excludes shape functions, makes the impact of crack depth
and location on flexural deformation clearer, aiding in crack identification. The developed
matrix is ideal for modeling flexural cracks in beams and columns, which is relevant in
earthquake engineering per European design code EC8. Roque [32] explored the symbolic and
numerical analysis of bending plates using MATLAB for symbolic manipulation of
expressions. The author emphasized the importance of integrating numerical and symbolic
approaches in problem-solving, highlighting MATLAB’s versatility in seamlessly combining
these computational methods.

1.2 Advantages of Symbolic Expressions and Solutions in MSA

The symbolic representation of structural analysis problems provides engineers and
researchers with deeper insight into structural behavior. By maintaining algebraic relationships
between parameters, symbolic MSA enables the exploration of how changes in material
properties, geometry, or loading affect the overall response of the structure. Symbolic solutions
are particularly useful in scenarios where flexibility and adaptability are crucial, as they provide
parameterized solutions not confined to specific input values. Furthermore, when symbolic
expressions are available, engineers can easily calculate partial derivatives with respect to
various input parameters, facilitating sensitivity analyses and design optimization in a
straightforward and efficient manner.
An additional advantage of symbolic MSA lies in its educational value. For teaching
structural engineering concepts, symbolic solutions can help students and professionals better
understand the fundamental principles of mechanics and structural behavior. The symbolic
form explicitly reveals how physical parameters such as the modulus of elasticity or the moment
of inertia of the beam’s cross-section influence displacements, stresses, and strains. This level
of transparency fosters a deeper conceptual understanding compared to purely numerical

4
 Vagelis Plevris and Afaq Ahmad

approaches, which often provide results without revealing the underlying mechanics.
Consequently, symbolic MSA serves as both a powerful research tool and an effective
educational resource for conveying complex ideas in a clear and accessible way.

1.3 Key Features and Novelty of this Study

This study presents an innovative, open-source MATLAB program specifically designed for
symbolic MSA of continuous beams subjected to point and uniform loads. For the first time,
this program enables the accurate and efficient derivation of analytical solutions for any
continuous beam configuration, providing engineers with symbolic expressions for
displacements, support reactions, and internal forces. These capabilities surpass traditional
methods by offering solutions that can be adapted to any beam setup with ease.
Users can define problems to obtain analytical solutions for specific output quantities at any
point x along the beam or determine the maximum values of these quantities. The program also
facilitates the generation of influence lines through symbolic expressions and supports
sensitivity analysis by calculating partial derivatives of outputs with respect to input parameters.
The main features of the program include:
• Closed-form solutions for any output quantities (e.g., support reactions) across
various continuous beam configurations.
• Solutions expressed as functions of x for specific points on the beam (e.g., bending
moment M(x)) and determination of maximum values along the beam (e.g., Mmax).
• Analytical expressions for influence lines for specified outputs (e.g., support
reactions influenced by a unit load moving along the beam).
• Sensitivity analysis for any output quantity relative to input parameters, utilizing
MATLAB’s built-in differentiation capabilities.

The source code is freely available at GitHub (https://github.com/vplevris/SymbolicMSA-

Beams), accompanied by all examples discussed in the study. This work was implemented
using MATLAB R2024b and its symbolic toolbox, and it is anticipated to be compatible with
earlier versions. Users are encouraged to download the code, experiment, and create their own
analytical solutions. The implementation is user-friendly, requiring minimal setup to execute
custom analyses.
2 Continuous Beam Stiffness Matrix
The stiffness matrix plays a crucial role in MSA as it defines the relationship between the
applied forces and the resulting displacements within a structural system. Typically, in
MSA/FEM stiffness matrices are computed numerically for each individual element and
subsequently assembled into a global system of equations representing the entire structure.
Numerical integration is usually needed, for example for the case of 2D plane stress elements
or more complex elements, which is regularly done using Gauss quadrature. However, in MSA,
for linear-type elements, it is usually possible to derive an exact symbolic expression for the
stiffness matrix of an element. One such case is the 2D Euler–Bernoulli beam with 3 degrees

5
 Deriving Analytical Solutions Using Symbolic Matrix Structural Analysis: Part 1 – Continuous Beams

of freedom (DOFs) per node [15] which is commonly used in the linear static analysis of plane
beams and frames. In this study, we use a beam element with 2 DOFs per node, one translational
and one rotational, resulting in 4 DOFs for the element, as shown in Figure 1. The axial degree
of freedom is omitted because the specific element is primarily designed for the analysis of
continuous beams, where axial effects are typically negligible.
y

1 3

x
i j
L

Figure 1. 2D Euler–Bernoulli beam element with 4 DOFs.

In this context, calculating the element stiffness matrix is straightforward. The symbolic 4×4
stiffness matrix for the Euler-Bernoulli beam element, assuming the axial degree of freedom is
neglected, corresponding to the DOFs shown in Figure 1, is given by Eq. (1).
 12 6 L −12 6 L 
EI  6 L 4 L −6 L 2 L 
2 2
[kˆb 2 ] = 3   (1)
L  −12 −6 L 12 −6 L 
 2 
 6 L 2 L −6 L 4 L 
2

In this expression, E denotes Young’s modulus of the material, I represents the moment of
inertia of the beam’s cross-section, and L is the length of the beam element. For continuous
beam analysis, where all beam elements are essentially horizontal and oriented from the left
(Start node i) to the right (End node j), the local stiffness matrix aligns with the global stiffness
matrix of the element. This eliminates the need for any element transformation, as the rotation
angle of the element is zero.
3 Educational Benefits and Importance of Symbolic Solutions in Structural Analysis
Symbolic solutions in the context of matrix structural analysis offer unique educational
advantages that go beyond the limitations of traditional numerical approaches. By retaining key
parameters in symbolic form, symbolic MSA enables students and engineers to explore the
fundamental relationships that govern structural behavior. This approach provides a more
intuitive understanding of how structural systems respond to changes in their properties, making
it an invaluable tool for teaching and reinforcing concepts in structural mechanics.
One of the primary benefits of using symbolic solutions in structural analysis is the ability
to visualize how different parameters affect the overall structural response. In a traditional
numerical MSA solution, students are often provided with final displacement values, reaction
forces, or internal stresses without a clear connection to the underlying properties of the
structure. However, in symbolic form, the solutions remain expressed in terms of parameters

6
 Vagelis Plevris and Afaq Ahmad

like E, I, and L, allowing students to directly observe how these variables influence the results.
For example, the symbolic stiffness matrix for a continuous beam of Eq. (1), shows how
stiffness increases with higher values of E and I, or how it decreases as the span length L
increases. This direct relationship between structural parameters and responses provides
students and engineers with a much deeper conceptual understanding of the mechanics
involved.
Another illustrative example is a simply supported horizontal beam subjected to a vertical
point load P at midspan. A purely numerical solution would provide a displacement value at
the midpoint, specific to the given load and beam properties. In contrast, a symbolic solution
will express the displacement at midspan as:
PL3
U= (2)
48EI
In this symbolic expression, students can immediately see the direct dependence of
displacement on the applied load P, the span length L, and the material and geometric properties
E and I. For instance, increasing the span length or decreasing the material stiffness results in a
larger displacement. Students can also realize that doubling L leads to an eightfold increase in
displacement, which makes L an important parameter for the displacement. The symbolic form
not only reinforces the theory behind structural deflection but also provides a framework for
students to experiment with different values of these parameters and predict how changes will
affect the behavior of the structure.
Symbolic solutions are also beneficial in helping students grasp the concept of
superposition of effects. For instance, in the analysis of continuous beams with multiple spans
and varying loads, symbolic expressions for reactions and internal forces can be developed for
each span or section of the beam. By analyzing these expressions, users can see how different
loads contribute to the overall structural response. The clarity of these symbolic relationships
allows for a step-by-step breakdown of how each load influences the displacement and internal
forces, which can be less apparent in numerical MSA results.
Another key advantage of symbolic MSA is its ability to enhance the teaching of design
sensitivity and optimization. In a classroom setting, students are often asked to investigate
how sensitive a structure’s response is to changes in material properties or geometry. Symbolic
solutions provide an ideal platform for this kind of analysis. For example, by differentiating a
symbolic expression for displacement with respect to Young’s modulus E, students can see how
changes in material stiffness will affect the structure’s performance. This provides an intuitive
understanding of the sensitivity of structural response to design changes, which is crucial for
both design optimization and practical engineering applications. Sensitivity analysis will be
examined in detail in Section 6.
In comparison to purely numerical solutions, symbolic expressions have the added
pedagogical benefit of clarity. Numerical MSA solutions often provide precise but isolated
results that do not offer insight into the overall behavior of the system. These results can obscure
key concepts, especially for students who are new to MSA. By contrast, symbolic MSA
maintains a clear and general form that highlights the relationships between different aspects
of the system, such as material properties, geometry, and loading conditions. This clarity allows

7
 Deriving Analytical Solutions Using Symbolic Matrix Structural Analysis: Part 1 – Continuous Beams

students to see not just the solution but also the process by which the solution is reached. This
is critical for building a solid theoretical foundation and developing problem-solving skills.
In summary, symbolic solutions in MSA provide significant educational benefits by making
the relationships between structural parameters and responses more transparent and accessible.
The use of symbolic expressions helps students visualize and understand how changes in
properties like material stiffness, geometry, or loading affect the structural response, reinforcing
theoretical concepts that might otherwise be obscured in purely numerical solutions. This
approach not only enhances learning but also equips students with the skills to perform
parametric studies, sensitivity analyses, and optimization in practical engineering contexts. As
a result, symbolic MSA is an invaluable tool for educators and students alike in mastering the
fundamentals of structural mechanics and matrix structural analysis.
4 Symbolic Expressions for Intermediate Values of Internal Forces and Local
Extrema
In matrix structural analysis of beams, whether conducted numerically or symbolically, the
results typically include displacements (or rotations) at nodal points and element forces at the
ends of each element (nodes i and j). While intermediate values for displacements or internal
forces are not provided directly by the method, these quantities can be determined indirectly by
applying established principles of statics.
It is quite straightforward to find the intermediate values of internal forces (shear force and
bending moment) at any given point x within an element, based on the known forces at the ends.
We consider the general case of a beam element with uniform load q on it (positive q points
upwards, towards the y axis of the beam), as shown in Figure 4.
y

Vi q Vj

x
i j
L

Figure 2. 2D beam element with uniform load q on.

The formulas for the internal shear force and the bending moment at a given location, x,
calculated from the left side, are:
V ( x) = Vi + q  x (3)

 x2  x2
M ( x) = −  M i − Vi  x − q  = q + Vi  x − M i (4)
 2 2
Here, x represents the distance along the local element x-axis, measured from the left node
(Node i) of the element. When x=0, it corresponds to the starting point of the element (Node i),

8
 Vagelis Plevris and Afaq Ahmad

while x=L marks the end of the element (Node j). The end moments, Mi and Mj, are defined as
positive when they act counter-clockwise, consistent with the notation in Figure 1. On the other
hand, for the shear force and bending moment diagrams, the internal shear force V(x) is positive
when it induces clockwise rotation of the element, and the internal bending moment M(x) is
considered positive when it induces tension in the bottom fiber of the beam. These conventions
are detailed in Figure 3(a) and Figure 3(b). In Eq. (4), the internal bending moment is calculated
from the left side, necessitating a negative sign to align with this convention and ensure
positivity when it induces tension in the bottom fiber, as illustrated in Figure 3(b).
Top Top

Vi Beam
Vj V + V

i Bottom
j Bottom

(a) (b)

Figure 3. Positive notation used (a) For the MSA results at each end of the beam (i and j),
(b) For the shear force and bending moment diagrams.

For the case of a uniform load (q≠0), the extremum (maximum or minimum) bending
moment is typically found at the point where the shear force equals zero, i.e., at
Vi
xextrM = − (5)
q
And according to Eq. (4) the value of the extremum bending moment will then be
2
 Vi 
− q 
 Vi  2
M ( xextrM ) = − M i + Vi   −  + w   = − M − Vi
i (6)
 q 2 2q
5 Numerical Examples
We consider five numerical examples of differing levels of complexity. In these, EI is treated
as a single symbolic parameter, since E and I consistently appear together in stiffness and other
analytical expressions. The first three examples have only point loads, while the other two have
uniform loads. Table 1 and Table 2 provide detailed descriptions of the five examples and the
associated symbolic parameters for each.

9
 Deriving Analytical Solutions Using Symbolic Matrix Structural Analysis: Part 1 – Continuous Beams

Table 1. Details of the numerical examples 1, 2, and 3.

Example # Figure Symbolic Parameters
P

3
1
(EI, L, and P)
0.8L 0.2L
L
P

4
2
(EI, L, a, and P)
a L-a
L
P1 P2

5
3
(EI, a, b, P1, and P2)
a a b b
L1=2a L2=2b

Table 2. Details of the numerical examples 4 and 5.

Example # Figure Symbolic Parameters
w

5
4
(EI, L, a, b, and w)
a b L-a-b
L
w
3
5
(EI, L, and w)
L L L L

5.1 Simply Supported Beam with a Point Load (3 Symbolic Parameters)

The first numerical example is a simply supported beam with a point load P applied at x=0.8L,
as shown in Figure 4. The symbolic parameters are three: EI, L, and P.

10
 Vagelis Plevris and Afaq Ahmad

0.8L 0.2L
L
Figure 4. The beam of Example 1.

Table 3 shows the details of the model, as given in MATLAB. In every example, the number
of elements (NumElements) is defined by the size of the Lengths vector and the number of nodes
(NumNodes) is (NumElements-1). In this example we have 2 elements and 3 nodes. Note that a
node is needed to define any point load.

Table 3. Details of the input parameters of the 1st numerical example.

0.8  L 0.2  L
T
Lengths
T
1 0 1 
Supports 0 0 0 
 
0 − P 0 
T

PointLoads 0 0 0 
 
0 0
T
UniformLoads

Table 4, Table 5 and Table 6 present the results of the symbolic analysis in terms of the
symbolic parameters. Table 4 presents the Node displacements and rotations, while Table 5
shows the support reactions and Table 6 the element forces and moments at the start (i) and end
(j) of each element.
Table 4. Example 1: Node displacements and rotations.

Node # y-Displacement (Dy) z-Rotation (Rz)

4 PL2
Node 1 0 −
125EI
16 PL3 4 PL2
Node 2 −
1875 EI 125 EI
6 PL2
Node 3 0
125 EI

Table 5. Example 1: Support reactions.

Node # Force Fy Moment Mz

Node 1 P5 -
Node 3 4P 5 -

11
 Deriving Analytical Solutions Using Symbolic Matrix Structural Analysis: Part 1 – Continuous Beams

Table 6. Example 1: Element forces and bending moments.

Element # Start / End Shear Force (Vi, Vj) Moment (Mi, Mj)
Start (i) P5 0
1 −P 5
End (j) 4PL 25
Start (i) −4 P 5 −4PL 25
2
End (j) 4P 5 0

5.2 Fixed-End Beam with Point Load (4 Symbolic Parameters)

The second numerical example is a fixed-end beam with a point load P, as shown in Figure 5.
The symbolic parameters are four: EI, L, x, and P. This example will also serve to illustrate the
concept of influence lines.

x L-x

Figure 5. The beam of Example 2.

Table 7 shows the details of the model, as given in MATLAB. In this example we have 3 nodes
and 2 elements.

Table 7. Details of the input parameters of the 2nd numerical example.

x L − x
T
Lengths
T
1 0 1
Supports 0 0 1
 
0 − P 0 
T

PointLoads 0 0 0 
 
0 0
T
UniformLoads

The results of the symbolic analysis are given in Table 8, Table 9, and Table 10, for Node
displacements and rotations; support reactions; and element forces and moments, respectively.

12
 Vagelis Plevris and Afaq Ahmad

Table 8. Example 2: Node displacements and rotations.

Node # y-Displacement (Dy) z-Rotation (Rz)

Px( L − x)2
Node 1 0 −
4 EIL
Px2 ( L − x)3 (3L + x) Px( L − x) (− L2 + 2Lx + x 2 )
2
Node 2 −
12EIL3 4EIL3
Node 3 0 0

Table 9. Example 2: Support reactions.

Node # Force Fy Moment Mz

P( L − x) (2 L + x)
2
Node 1 -
2 L3
Px(3L2 − x 2 ) Px( L2 − x 2 )
Node 3 −
2L3 2L2

Table 10. Example 2: Element forces and bending moments.

Element # Start / End Shear Force (Vi, Vj) Moment (Mi, Mj)
P( L − x) (2 L + x)
2
Start (i) 0
2 L3
1
P( L − x)2 (2L + x) Px( L − x)2 (2L + x)
End (j) −
2L3 2L3
Px(3L2 − x 2 ) Px( L − x)2 (2L + x)
Start (i) − −
2 L3 2L3
2
Px(3L2 − x 2 ) Px( L2 − x 2 )
End (j) −
2L3 2L2

Now suppose we would like to draw the influence line of the reaction moment at Node 3.
According to the results shown in Table 9, the reaction moment at Node 3 (counter-clockwise
is positive) is given by
Px( L2 − x 2 )
M z ,3 = − (7)
2 L2
To draw the influence line, we assume a unit load (P=1) applied along the beam. For
illustration purposes, we will assume that the length of the beam is also unitary (L=1). Then the
reaction moment is given by:
x(1 − x 2 )
M z ,3 = − (8)
2
The influence line corresponding to Eq. (8) is presented in Figure 6.

13
 Deriving Analytical Solutions Using Symbolic Matrix Structural Analysis: Part 1 – Continuous Beams

Figure 6. The influence line of the reaction moment at Node 3, for Example 3.

We observe that the reaction moment takes the value of zero when the load is at the
beginning (x=0) or the end (x=L) of the beam, while it takes its maximum (in absolute terms)
value M max = − 3 9  −0.1925 when the unitary load is at x = 3 3  0.5774 .

5.3 Two-Span Beam with Two Point Loads (5 Symbolic Parameters)

The third numerical example is a two-span beam with two point loads, as shown in Figure 7.
The symbolic parameters are six: EI, a, b, P1, and P2.

P1 P2

a a b b
L1=2a L2=2b
Figure 7. The two-span beam of Example 3.

Table 11 shows the details of the model, as given in MATLAB.

14
 Vagelis Plevris and Afaq Ahmad

Table 11. Details of the input parameters of the 3rd numerical example.

a a b b
T
Lengths
T
1 0 1 0 1 
Supports 0 0 0 0 0 
 
0 − P1 0 − P2
T
0
PointLoads 0 0 0 0 0 

0 0 0 0
T
UniformLoads

The results of the symbolic analysis are reported in Table 12,

Table 13, and Table 14, for Node displacements and rotations; support reactions; and element
forces and moments, respectively.

Table 12. Example 3: Node displacements and support reactions.

Node # y-Displacement (Dy) z-Rotation (Rz)

a( Pa 2
+ 2 Pab − P2b 2 )
Node 1 0 − 1 1

8EI (a + b)
a 2 (7 Pa 2
+ 16 Pab − 9 P2b 2 ) a( Pa 2
+ P2b 2 )
Node 2 − 1 1 1

96 EI (a + b) 32 EI (a + b)
ab ( Pa
1 − P2 b )
Node 3 0 4 EI (a + b)
b (−9 Pa
2
+ 16 P2 ab + 7 P2b )
2 2
b( Pa 2
+ P2b 2 )
Node 4 − 1
− 1

96 EI (a + b) 32 EI (a + b)
b(− Pa
1
2
+ 2 P2 ab + P2b 2 )
Node 5 0 8EI (a + b)

Table 13. Example 3: Support reactions.

Node # Force Fy Moment Mz

5Pa
1 + 8Pab
2
1 − 3P2b 2

Node 1 -
16a(a + b)
3Pa
1
2
+ 3P2 b 2 + 8Pab
1 + 8P2 ab
Node 3 -
16ab
−3Pa1
2
+ 8P2 ab + 5P2 b 2
Node 5 16b(a + b)
-

15
 Deriving Analytical Solutions Using Symbolic Matrix Structural Analysis: Part 1 – Continuous Beams

Table 14. Example 3: Element forces and bending moments.

Element # Start / End Shear Force (Vi, Vj) Moment (Mi, Mj)
5Pa
1 + 8Pab
2
1 − 3P2b 2

Start (i) 0
16a(a + b)
1
5Pa 2
+ 8Pab − 3P2 b 2 5Pa 2
+ 8Pab − 3P2b 2
End (j) − 1 1 1 1

16a(a + b) 16(a + b)
11Pa 2
+ 8Pab + 3P2b 2 5Pa 2
+ 8Pab − 3P2 b 2
Start (i) − 1 1
− 1 1

16a(a + b) 16(a + b)
2
11Pa 2
+ 8Pab + 3P2 b 2 3( Pa 2
+ P2b 2 )
End (j) 1 1
− 1

16a(a + b) 8(a + b)
3Pa
1
2
+ 8P2 ab + 11P2 b 2 3( Pa
1
2
+ P2 b 2 )
Start (i)
16b(a + b) 8(a + b)
3
3Pa 2
+ 8P2 ab + 11P2b 2 −3Pa 2
+ 8P2 ab + 5P2 b 2
End (j) − 1 1

16b(a + b) 16(a + b)
−3Pa 2
+ 8P2 ab + 5P2 b 2 −3Pa 2
+ 8P2 ab + 5P2 b 2
Start (i) − 1
− 1

16b(a + b) 16(a + b)
4
−3Pa
1
2
+ 8P2 ab + 5P2 b 2
End (j) 0
16b(a + b)

5.4 Simply Supported Beam with Uniform Load (5 Symbolic Parameters)

The fourth numerical example is a simply supported beam with a uniform load w, as shown in
Figure 8. The symbolic parameters are six: EI, L, a, b, and w.
w

a b L-a-b
L
Figure 8. The beam of Example 4.

Table 15 shows the details of the model, as given in MATLAB. In this example we have 4
nodes and 3 elements.

16
 Vagelis Plevris and Afaq Ahmad

Table 15. Details of the input parameters of the 4th numerical example.

a b L − a − b
T
Lengths
T
1 0 0 1 
Supports 0 0 0 0
 
T
0 0 0 0
PointLoads 0 0 0 0
 
0 − w 0
T
UniformLoads
The results of the symbolic analysis are given in Table 16, Table 17, and Table 18, for Node
displacements and rotations; support reactions; and element forces and moments, respectively.

Table 16. Example 4: Node displacements and rotations.

Node # y-Displacement (Dy) z-Rotation (Rz)

bw(8L2 a + 4L2b − 12La 2 − 12Lab − 4Lb2 + 4a3 + 6a 2b + 4ab2 + b3 )
Node 1 0 −
24EIL
abw(8L2 a + 4L2b − 16La 2 − 12Lab − 4Lb 2 + 8a3 + 8a 2b + 4ab 2 + b3 ) bw(8L2 a + 4L2b − 24La 2 − 12Lab − 4Lb2 + 16a3 + 12a 2b + 4ab2 + b3 )
Node 2 − −
24EIL 24EIL

bw(2a + b)(a − L + b)(4a 2 + 6ab − 4La + 3b2 − 4Lb) bw(2a + b)(4L2 − 12La − 12Lb + 8a 2 + 14ab + 7b2 )
Node 3 − −
24EIL
24EIL
bw(2a + b)(−2L2 + 2a 2 + 2ab + b2 )
Node 4 0 −
24EIL

Table 17. Example 4: Support reactions.

Node # Force Fy Moment Mz

bw ( 2a − 2L + b )
Node 1 − -
2L
bw ( 2a + b )
Node 4 -
2L

Table 18. Example 4: Element forces and bending moments.

Element # Start / End Shear Force (Vi, Vj) Moment (Mi, Mj)
bw(2a − 2 L + b)
Start (i) − 0
2L
1
bw(2a − 2 L + b) abw(2a − 2 L + b)
End (j) −
2L 2L
bw(2a − 2 L + b) abw(2a − 2 L + b)
Start (i) −
2L 2L
2
bw(2a + b) bw ( 2a + b )( a − L + b )
End (j) −
2L 2L
bw(2a + b) bw ( 2a + b )( a − L + b )
Start (i) −
2L 2L
3
bw(2a + b)
End (j) 0
2L

17
 Deriving Analytical Solutions Using Symbolic Matrix Structural Analysis: Part 1 – Continuous Beams

The first three numerical examples did not include any uniform loads. In this fourth example,
by applying the formulas of Eqs. (5) and (6) for Element 2 (with uniform load q=-w), we get:
2a − 2 L + b
xextrM = −b  (9)
2L
b 2 w(2a − 2 L + b) 2 abw(2a − 2 L + b)
M ( xextrM ) = − (10)
8L2 2L
We see that we get the exact value of the maximum bending moment along Element 2, in a
symbolic form. By combining the results presented in Table 18 and Eqs. (9) and (10), we can
draw the bending moment diagram of the beam in a symbolic way, including the maximum
bending moment at the span 2, as shown in Figure 9.
b 2 w(2a − 2 L + b) 2 abw(2a − 2 L + b)
− bw ( 2a + b )( a − L + b )
8L2 2L −
abw(2a − 2 L + b) 2L
−
2L

Figure 9. Symbolic bending moment diagram for Example 4.

5.5 Four-Span Beam with Uniform Load (3 Symbolic Parameters)

The fifth numerical example is a four-span continuous beam with a uniform load w, fixed at its
right end, where all spans have the same length L, as shown in Figure 10. Although the
computational model is larger and more complex than the previous ones and there are more
nodes and elements, the symbolic parameters in this example are only three: EI, L, and w.

L L L L
Figure 10. The four-span beam of Example 5.

Table 19 shows the details of the model, as given in MATLAB. In this example we have 5
nodes and 4 elements.

18
 Vagelis Plevris and Afaq Ahmad

Table 19. Details of the input parameters of the 5th numerical example.

L L L L
T
Lengths
T
1 1 1 1 1
Supports 0 0 0 0 1
 
T
0 0 0 0 0 
PointLoads 0 0 0 0 0 
 
 −w −w −w −w
T
UniformLoads

The results of the symbolic analysis are given in Table 20, Table 21, and Table 22, for Node
displacements and rotations; support reactions; and element forces and moments, respectively.

Table 20. Example 5: Node displacements and rotations.

Node # y-Displacement (Dy) z-Rotation (Rz)

7 wL3
Node 1 0 −
291EI
5wL3
Node 2 0
776 EI
wL3
Node 3 0 −
582 EI
wL3
Node 4 0
2328 EI
Node 5 0 0

Table 21. Example 5: Support reactions.

Node # Force Fy Moment Mz

153wL
Node 1 -
388
110wL
Node 2 -
97
187 wL
Node 3 -
194
98wL
Node 4 -
97
193wL 8wL2
Node 5 −
388 97

19
 Deriving Analytical Solutions Using Symbolic Matrix Structural Analysis: Part 1 – Continuous Beams

Table 22. Example 5: Element forces and bending moments.

Element # Start / End Shear Force (Vi, Vj) Moment (Mi, Mj)
153wL
Start (i) 0
388
1
235wL 41wL2
End (j) −  −0.105670wL2
388 388
205wL 41wL2
Start (i)  −0.105670wL2
388 388
2
183wL 15wL2
End (j) −  −0.077320 wL2
388 194
191wL 15wL2
Start (i)  −0.077320wL2
388 194
3
197 wL 33wL2
End (j) −  −0.085052wL2
388 388
195wL 33wL2
Start (i)  −0.085052wL2
388 388
4
193wL 8wL2
End (j) −  −0.082474 wL2
388 97

In this example, by applying the formulas of Eqs. (5) and (6) for all elements (as all elements
have a uniform load q=-w on them), we get the results presented in Table 23 for the maximum
value of the internal bending moment for each span.

Table 23. Location and value of max. bending moment for each span of the beam (5th numerical example).

Element # xextrM M(xextrM)

2
153L 23409wL
1  0.077748wL2 *
388 301088
205L 10209wL2
2  0.033907 wL2
388 301088
191L 13201wL2
3  0.043844 wL2
388 301088
195 L 12417wL2
4  0.041240 wL2
388 301088
* Max value at any span.
By taking the results presented in Table 22 (for element end values) and Table 23 (for span
values), we can draw the bending moment diagram of the beam in a symbolic way, as shown
in Figure 9.

20
 Vagelis Plevris and Afaq Ahmad

0.077748wL2

0.043844wL2 0.041240wL2
0.033907wL2

−0.077320wL2
−0.085052wL2 −0.082474wL2

−0.105670wL2

Figure 11. Symbolic bending moment diagram for Example 5.

6 Sensitivity Analysis
Sensitivity analysis is a powerful technique used in structural engineering to evaluate how
variations in input parameters affect the system’s response. It plays a critical role in
understanding the robustness of a design and in optimizing performance by identifying which
parameters have the most significant impact on the system’s behavior. In structural analysis,
engineers may want to know how changes in material properties, geometric dimensions, or
applied loads affect displacements, stresses, reaction forces, or internal moments. Sensitivity
analysis provides insight into these relationships, enabling more informed decision-making
during the design process.
One of the major advantages of using symbolic solutions for structural engineering problems
is that sensitivity analysis becomes remarkably straightforward. When the behavior of a
structural system is expressed symbolically, the engineer has access to closed-form expressions
that describe how key parameters such as beam length, Young’s modulus, or applied loads
influence the system’s response. This allows for the direct calculation of partial derivatives of
any output with respect to any input parameter. For example, with a symbolic expression for
displacement, one can easily compute the sensitivity of the displacement at a specific point to
changes in beam length. Similarly, the sensitivity of the internal bending moment at a given
point, or a support reaction to variations in load or material properties can be computed.
For instance, consider the second numerical example shown in the previous section. The
vertical displacement at x=a, where the load P is applied, according to Table 8 is given by
Px 2 ( L − x)3 (3L + x)
Dy , x =− (11)
12 EIL3
Using the MATLAB Symbolic Math Toolbox, we can compute the partial derivative of the
displacement with respect to any parameter (P, x, L, EI), giving us a mathematical expression
for how changes of a parameter affect the displacement. In this example, the partial derivative

21
 Deriving Analytical Solutions Using Symbolic Matrix Structural Analysis: Part 1 – Continuous Beams

of Dy,x with respect to L can be found with the command “simplify(diff(Dyx, L))”, where the
MATLAB command diff is used for symbolic differentiation and the command simplify is used
to simplify the expression as much as possible. The result is the following:
Dy , x Px 2 ( L2 − x 2 ) 2
=− (12)
L 4 EIL4
This partial derivative provides the sensitivity of the displacement to changes in beam
length, offering valuable insight into how modifications in the design will affect the
performance of the structure. The same approach can be applied to assess the sensitivity of
internal forces, bending moments, or support reactions to variations in parameters like cross-
sectional moment of inertia, material stiffness, or applied loads.
This ability to differentiate symbolic expressions provides a significant advantage over
numerical solutions. In numerical analysis, we only obtain specific results for specific input
values, making it difficult or impossible to predict how small changes in parameters will affect
the outcome without rerunning the entire analysis for each scenario. Symbolic expressions, on
the other hand, provide a general solution that retains the relationship between inputs and
outputs, allowing for easy exploration of parameter sensitivities without additional
computational cost.
Performing these calculations symbolically gives engineers an extra layer of insight,
enabling them not only to see the results of an analysis but also to predict how different input
parameters will influence the outputs. This ability to quickly and accurately evaluate the effects
of parameter changes is especially important in the optimization phase of design, where small
adjustments can significantly impact the overall performance and cost-effectiveness of a
structure. This capability, which is unique to symbolic expressions, empowers engineers to
predict the effects of design changes without the need for repeated numerical analysis, saving
time and providing a clearer understanding of the system’s behavior.
7 Efficiency, Scalability and Elegance of Symbolic Computations
While the symbolic implementation of MSA can offer significant advantages in terms of
flexibility and insight, it also presents challenges in terms of computational efficiency and
scalability, particularly for larger systems. As structural systems grow in complexity—whether
through increased degrees of freedom, more elements, or complex loading conditions—the
symbolic representations of stiffness matrices, force vectors, and displacement fields can
become increasingly large and computationally demanding. It is important to address these
challenges to ensure that symbolic MSA remains practical and efficient for engineering
applications.
One of the primary issues with symbolic computations is the exponential growth in the size
of symbolic expressions as the complexity of the problem increases. For a simple system, a
symbolic stiffness matrix or equilibrium equation may provide a clear and elegant solution.
However, for larger systems or those involving a large number of symbolic variables, the
resulting expressions can quickly become excessively long and complex, making them difficult
to manage, while losing their interpretability and usefulness for the engineer. For instance, a
symbolic solution that spans multiple pages with dense expressions may offer little practical

22
 Vagelis Plevris and Afaq Ahmad

value compared to a simple numerical result. This is a critical limitation because, even though
symbolic software like MATLAB will always attempt to generate a symbolic solution, the
elegance and simplicity of the solution are paramount for engineers. A symbolic expression
that is excessively long and convoluted is not only difficult to work with but also loses its value
in providing insight into the system’s behavior.
To manage this complexity, one effective approach is to limit the number of symbolic
variables used in the formulation. For instance, rather than expressing every parameter of the
system symbolically, it may be beneficial to identify key parameters that are most important to
the analysis. By reducing the number of symbolic variables to a smaller, more manageable set,
the symbolic representation remains compact and comprehensible, even for larger systems. This
hybrid symbolic-numerical approach allows for symbolic manipulation of the most
important parameters while handling the less critical parts of the system numerically. This
combination of symbolic and numerical methods provides a balance between flexibility and
computational efficiency, making it possible to apply symbolic MSA to larger or more complex
systems without overwhelming the computational resources. As a result, the solution retains its
generality and flexibility without sacrificing efficiency.
The elegance of the solution is what ultimately matters to the engineer. If the symbolic
solution is compact, easy to interpret, and provides useful insights, it can be extremely valuable.
On the other hand, if the solution is overly large and difficult to work with, it loses its practical
utility, even if it technically provides a valid answer. Striking the right balance between
symbolic and numerical methods ensures that engineers can take advantage of the benefits of
symbolic computation without being overwhelmed by computational complexity.
8 Conclusions
This study has presented the development and application of an open-source MATLAB
program capable of performing symbolic matrix structural analysis for continuous beams under
point and uniform loads. This tool is freely available and provides the capability to rapidly and
accurately generate analytical solutions for any beam configuration. The profound benefits of
obtaining analytical solutions extend beyond practical engineering applications to enhancing
engineering education, where such solutions help deepen the understanding of structural
behavior.
The code goes beyond simply deriving analytical solutions for displacements, support
reactions, and internal forces. It can also be used to produce influence lines for continuous
beams and facilitate sensitivity analysis, which is vital for optimization and other engineering
applications. The ability to compute partial derivatives of output parameters with respect to
input variables enables users to assess how changes in design properties influence the overall
response, making the tool valuable for both design exploration and performance evaluation.
While symbolic solutions offer significant advantages, it is essential to approach them with
consideration for elegance and simplicity. The program is capable of producing large and
complex symbolic expressions, but these may lack practical usefulness if they become too
intricate to interpret or apply effectively. On the other hand, concise and simpler analytical
solutions fulfill their intended purpose by being more comprehensible and actionable. The
efficiency, scalability, and elegance of symbolic computations are crucial aspects that engineers

23
 Deriving Analytical Solutions Using Symbolic Matrix Structural Analysis: Part 1 – Continuous Beams

and students must balance when employing this approach.

Through various examples, we have demonstrated the power of the code and showcased its
potential to offer insightful analytical solutions. The next step in this line of research is to extend
the methodology to other structural systems, such as 2D trusses and frames, to derive symbolic
solutions for these structures as well. This future research direction holds promise for expanding
the application of symbolic MSA and enhancing its contribution to structural analysis and
education.
Conflict of Interest
The authors declare that the research was conducted in the absence of any commercial or
financial relationships that could be construed as a potential conflict of interest.

References
[1] Liu, W.K., S. Li, and H.S. Park, Eighty Years of the Finite Element Method: Birth,
Evolution, and Future. Archives of Computational Methods in Engineering, 2022. 29(6):
p. 4431-4453 DOI: https://doi.org/10.1007/s11831-022-09740-9.
[2] Plevris, V. and G. Tsiatas, Computational Structural Engineering: Past Achievements and
Future Challenges. Frontiers in Built Environment, 2018. 4(21): p. 1-5 DOI:
https://doi.org/10.3389/fbuil.2018.00021.
[3] Mazumder, S., Chapter 1 - Introduction to Numerical Methods for Solving Differential
Equations, in Numerical Methods for Partial Differential Equations, S. Mazumder, Editor.
2016, Academic Press. p. 1-49. DOI https://doi.org/10.1016/B978-0-12-849894-1.00001-
9.
[4] Papadrakakis, M., N.D. Lagaros, Y. Tsompanakis, and V. Plevris, Large Scale Structural
Optimization: Computational Methods and Optimization Algorithms. Archives of
Computational Methods in Engineering, 2001. 8(3): p. 239-301 DOI:
https://doi.org/http://dx.doi.org/10.1007/BF02736645.
[5] Lagaros, N.D., V. Plevris, and N.A. Kallioras, The Mosaic of Metaheuristic Algorithms in
Structural Optimization. Archives of Computational Methods in Engineering, 2022. 29: p.
5457–5492 DOI: https://doi.org/10.1007/s11831-022-09773-0.
[6] Farahmand-Tabar, S. and K. Aghani, Chapter 2 - Matrix Analysis of Framed Structures,
in Practical Programming of Finite Element Procedures for Solids and Structures with
MATLAB®, S. Farahmand-Tabar and K. Aghani, Editors. 2024, Elsevier. p. 29-116. DOI
https://doi.org/10.1016/B978-0-443-15338-9.00007-1.
[7] Geisel, A. and F. Svaricek, A MATLAB Toolbox for Structural Analysis of Linear Systems.
IFAC-PapersOnLine, 2019. 52(17): p. 7-12 DOI:
https://doi.org/10.1016/j.ifacol.2019.11.018.
[8] Yang, B., Chapter 6 - Static analysis of constrained multispan beams, in Stress, Strain,
and Structural Dynamics (Second Edition), B. Yang, Editor. 2023, Academic Press. p. 183-
223. DOI https://doi.org/10.1016/B978-0-12-818563-6.00018-3.
[9] Papazafeiropoulos, G. and V. Plevris, OpenSeismoMatlab: New Features, Verification and
Charting Future Endeavors. Buildings, 2024. 14(1): p. 304 DOI:
https://doi.org/10.3390/buildings14010304.

24
 Vagelis Plevris and Afaq Ahmad

[10] Yang, B., Chapter 9 - Dynamics of particles and rigid bodies, in Stress, Strain, and
Structural Dynamics (Second Edition), B. Yang, Editor. 2023, Academic Press. p. 351-
430. DOI https://doi.org/10.1016/B978-0-12-818563-6.00012-2.
[11] Yang, B., Chapter 14 - Dynamic analysis of constrained, combined, and stepped beams,
in Stress, Strain, and Structural Dynamics (Second Edition), B. Yang, Editor. 2023,
Academic Press. p. 801-872. DOI https://doi.org/10.1016/B978-0-12-818563-6.00019-5.
[12] Plevris, V. and G. Solorzano, A Collection of 30 Multidimensional Functions for Global
Optimization Benchmarking. Data, 2022. 7(4): p. 52 DOI:
https://doi.org/10.3390/data7040046.
[13] Andreassen, E., A. Clausen, M. Schevenels, B.S. Lazarov, and O. Sigmund, Efficient
topology optimization in MATLAB using 88 lines of code. Structural and Multidisciplinary
Optimization, 2011. 43(1): p. 1-16 DOI: https://doi.org/https://doi.org/10.1007/s00158-
010-0594-7.
[14] Sigmund, O., A 99 line topology optimization code written in Matlab. Structural and
Multidisciplinary Optimization, 2001. 21(2): p. 120-127 DOI:
https://doi.org/10.1007/s001580050176.
[15] Ferreira, A.J.M. and N. Fantuzzi, Bernoulli 2D Frames, in MATLAB Codes for Finite
Element Analysis: Solids and Structures. 2020, Springer International Publishing: Cham.
p. 105-121. DOI https://doi.org/10.1007/978-3-030-47952-7_7.
[16] MathWorks, Symbolic Math Toolbox™ User's Guide. 2024 DOI.
[17] Attaway, S., Chapter 14 - Advanced Mathematics, in MATLAB (Fifth Edition), S. Attaway,
Editor. 2019, Butterworth-Heinemann. p. 527-581. DOI https://doi.org/10.1016/B978-0-
12-815479-3.00014-3.
[18] Valentine, D.T. and B.D. Hahn, Chapter 17 - Symbolics toolbox, in Essential MATLAB for
Engineers and Scientists (Eighth Edition), D.T. Valentine and B.D. Hahn, Editors. 2023,
Academic Press. p. 351-368. DOI https://doi.org/10.1016/B978-0-32-399548-1.00024-2.
[19] Asadi, F., Symbolic Calculations in MATLAB®, in Applied Numerical Analysis with
MATLAB®/Simulink®: For Engineers and Scientists, F. Asadi, Editor. 2023, Springer
Nature Switzerland: Cham. p. 55-71. DOI https://doi.org/10.1007/978-3-031-19366-8_2.
[20] Mureşan, M., Introduction to Mathematica® with Applications. 2017: Springer Cham,
ISBN: 978-3-319-52003-2.DOI: 10.1007/978-3-319-52003-2.
[21] Smith, E., Maple, in Introduction to the Tools of Scientific Computing. 2022, Springer
International Publishing: Cham. p. 219-244. DOI https://doi.org/10.1007/978-3-031-
16972-4_10.
[22] Cywiak, M. and D. Cywiak, SymPy, in Multi-Platform Graphics Programming with Kivy:
Basic Analytical Programming for 2D, 3D, and Stereoscopic Design. 2021, Apress:
Berkeley, CA. p. 173-190. DOI https://doi.org/10.1007/978-1-4842-7113-1_11.
[23] Lynch, S., A Tutorial Introduction to MATLAB and the Symbolic Math Toolbox, in
Dynamical Systems with Applications using MATLAB®. 2004, Birkhäuser Boston: Boston,
MA. p. 1-13. DOI https://doi.org/10.1007/978-0-8176-8156-2_1.
[24] Korncoff, A.R. and S.J. Fenves, Symbolic generation of finite element stiffness matrices.
Computers & Structures, 1979. 10(1): p. 119-124 DOI: https://doi.org/10.1016/0045-
7949(79)90078-6.
[25] Leff, L. and D.Y.Y. Yun, The symbolic finite element analysis system. Computers &
Structures, 1991. 41(2): p. 227-231 DOI: https://doi.org/10.1016/0045-7949(91)90426-M.

25
 Deriving Analytical Solutions Using Symbolic Matrix Structural Analysis: Part 1 – Continuous Beams

[26] Nagabhushanam, J., C.J. Srinivas, and G.H. Gaonkar, Symbolic generation of elemental
matrices for finite element analysis. Computers & Structures, 1992. 42(3): p. 375-380 DOI:
https://doi.org/10.1016/0045-7949(92)90033-V.
[27] Tummarakota, S. and J. Lieh, Symbolic Finite Element Modeling of Structural Systems.
Journal of Symbolic Computation, 1996. 22(1): p. 105-119 DOI:
https://doi.org/10.1006/jsco.1996.0043.
[28] Eriksson, A. and C. Pacoste, Symbolic software tools in the development of finite elements.
Computers & Structures, 1999. 72(4): p. 579-593 DOI: https://doi.org/10.1016/S0045-
7949(98)00333-2.
[29] Amberg, G., R. Tönhardt, and C. Winkler, Finite element simulations using symbolic
computing. Mathematics and Computers in Simulation, 1999. 49(4): p. 257-274 DOI:
https://doi.org/10.1016/S0378-4754(99)00054-3.
[30] Pavlović, M.N., Symbolic computation in structural engineering. Computers & Structures,
2003. 81(22): p. 2121-2136 DOI: https://doi.org/10.1016/S0045-7949(03)00286-4.
[31] Skrinar, M. and T. Pliberšek, On the derivation of symbolic form of stiffness matrix and
load vector of a beam with an arbitrary number of transverse cracks. Computational
Materials Science, 2012. 52(1): p. 253-260 DOI:
https://doi.org/10.1016/j.commatsci.2011.07.013.
[32] Roque, C.M.C., Symbolic and numerical analysis of plates in bending using Matlab.
Journal of Symbolic Computation, 2014. 61-62: p. 3-11 DOI:
https://doi.org/10.1016/j.jsc.2013.10.005.

26