ANALYTICAL AND NUMERICAL METHODS IN STRUCTURAL ENGINEERING Supported by

DESIGN PRACTICE: IMPORTANT NOTE FOR YOUNG ENGINEERS

1Agbakwuru Jasper A;*2Osuji Joshua N.
*1Departmentof Marine Engineering, Federal University of Petroleum Resources, Effurun, Delta State, Nigeria.
2Department of Transport Management Technology, Federal University of Technology Owerri, Imo State, Nigeria

*Corresponding Author Email: [email protected]; Tel: +2348106891639

Other Author Email: [email protected]; Tel: +2348037820628

Received: May 18, 2023 Accepted: July 10, 2023

Abstract
Engineering structural design activities today have been positively impacted by the availability and use of
numerous engineering software, especially the Finite Element Method (FEM) of the numerical solution. Young
engineers and associate engineers have found engineering design with software not only comfortable but
fascinating and easy. However, there exist some inconsistent information in some literatures concerning the
use of the FEM software application. This work presents a typical engineering design case in which the
numerical FEM and the analytical solutions are used and compared. The work re-iterates that FEM for standard
engineering software such as SolidWorks etc., when properly deployed yields exact results derived from
analytical methods. It is noted that the result does not depend on the structural design but can depend on the
model assumptions which in that case depend on the user and not the software itself. The paper also provides
caution, especially to young engineers and associate structural engineers on the use of FEM software for design
and other engineering computations.
Keywords: Finite Element Method, SolidWorks FEM, numerical methods, analytical methods, structural design, FEM
software, Factor of Safety, structures

Introduction a dynamic application factor. It is unclear in the work of

Engineering of today is fast growing and different software Giuseppe et al. (2019) the bases for the overestimation of
for the design and analysis of structures are presently being numerical analysis method or underestimation of the
used both in the industry and academia. Young Engineers numerical FEM method with respect to the true values.
especially are unable to appreciate the essence of analytical Karthik et al.(2017) discussed analytical and numerical
results and computations compared to numerical methods. design analysis of concentric tube heat exchangers. Their
Generally speaking, analytical methods of computation work showed that both numerical and analytical approaches
intend to give exact solutions. The numerical methods on the produce the same results. The implication according to
other hand intend to provide approximate results (as close as Karthik et al. (2017) is that the designer is at liberty to choose
possible to the exact solution). The strength of a numerical any convenient method. Kuang-Hua (2015) discussed design
solution is a measure of how close the results are to the exact optimization and suggests that the trends of the structural
solution. It is important to examine these two techniques in stress determined with the analytical methods and numerical
engineering practice to validate the understanding and needs. simulation methods were similar.

Nunez et al. (2012) employed the use of numerical and The literature review is inconsistent on the agreement of
analytical methods to calculate stresses on a pile-supported results of numerical and analytical approaches to design
embankment over soft clay. The results were compared with methods. Is this agreement dependent on structural design?
experimental tests and Nunez et al. (2012) found an Is it dependent on the overall underlining model
overestimation of the stresses. The errors were attributed to assumptions? While Karthik et al. (2017) and Kuang-Hua
assumptions on different models used in the work. Giuseppe (2015) firmed that both approaches produce similar results,
et al. (2019) examined the analytical and numerical Nunez et al., (2012) and Giuseppe et al. (2019) demonstrated
approaches for the design of stone pavers in urban shared overestimation and underestimation of the true results.
areas. The results produced using the analytical theory of
Westergaard were found to be higher than those obtained This paper attempts to investigate the use of numerical and
from a static Finite Element Method. Giuseppe et al. (2019) analytical methods in structural engineering design work. In
seemed to accept that the analytical method yields an this work, a simple underwater structural design is utilized
overestimation of the true values while on the other hand for comparative solutions of analytical and numerical
arguing that it is faster and cheaper than numerical FEM methods. This article will consider an acrylic sight glass
method. Giuseppe et al. (2019) therefore tended to find (20cm x 20cm x 20mm thick) installed on a wet bell
optimal balance between the two approaches by introducing operating at 30 meters underwater (see Figure 1).

FUW Trends in Science & Technology Journal, www.ftstjournal.com

e-ISSN: 24085162; p-ISSN: 20485170; August, 2023: Vol. 8 No. 2 pp. xxx – xxx 52
 Analytical and Numerical Methods in Structural Engineering Design Practice: Important Note for Young Engineers

Figure 1: The model wet bell and its acrylic glass

Numerical method: Exposed area of the sight glass to the pressure, A = 20cm x
Engineering material failure criterion, especially for ductile 20cm = 0.04m2
materials is commonly Von Mises (Giraldo-Londoño and Consider that the difference 𝑑𝐻 over the surface of the
Paulino, 2022). The presented structure in Figure 1 is a acrylic glass is negligibly small compared to water depth, the
simple loading but can be generalized into biaxial situation load could be considered distributed over the acrylic surface.
as described by Hearn (1999) such that The distributed load 𝐹𝑤 = 𝑃 × 𝐴 = 𝑃 ×
Von-Mises stress Equivalent, 𝜎𝑒 = √(𝜎1 2 + 𝜎2 2 − 𝜎1 ∗ 𝜎2 ) 𝑤𝑖𝑑𝑡ℎ 𝑎𝑐𝑟𝑜𝑠𝑠 𝑡ℎ𝑒 𝑔𝑙𝑎𝑠𝑠 ℎ𝑟𝑜𝑧𝑜𝑛𝑡𝑎𝑙𝑙𝑦 ×
𝜎1 = 𝑘𝜎2 𝑙𝑒𝑛𝑔𝑡ℎ 𝑎𝑙𝑜𝑛𝑔 𝑡ℎ𝑒 ℎ𝑒𝑖𝑔ℎ𝑡 𝑣𝑒𝑟𝑡𝑖𝑐𝑎𝑙𝑙𝑦 (𝑑𝐻)
2 𝜎1 2 𝜎1 𝑤𝑖𝑑𝑡ℎ 𝑎𝑐𝑟𝑜𝑠𝑠 𝑡ℎ𝑒 𝑔𝑙𝑎𝑠𝑠 ℎ𝑟𝑜𝑧𝑜𝑛𝑡𝑎𝑙𝑙𝑦 = 0.2m
𝜎𝑒 = √(𝜎1 + ( ) − 𝜎1 ∗ ) …………………….(1)
𝑘 𝑘 𝑙𝑒𝑛𝑔𝑡ℎ 𝑎𝑙𝑜𝑛𝑔 𝑡ℎ𝑒 ℎ𝑒𝑖𝑔ℎ𝑡 𝑣𝑒𝑟𝑡𝑖𝑐𝑎𝑙𝑙𝑦 (ranging from 0m to
Where 𝑘 is the stress concentrated factor, 0 ≤ 𝑘 ≤ 1.0. For 0.2m maximum) = 𝑑𝐻
this simple case, 𝑘 = 1. 𝐹𝑤 = 301657.5×0.2× (𝑑𝐻) = 60331 N/m
1.2 Numerical analysis The sectional free body diagram is shown in Figure 2 for the
Finite element method of numerical analysis is discussed. distributed load case.
The physical engineering component of a structure is
stimulated using the finite element method. A good number
of software presently use finite element method of numerical
analysis for structural designs. Example is the SolidWorks
software which is referred to as Infinite Element Analysis
meshed into structures divided into small finite elements
(Muhammad and Shanono, 2021). According to Muhammad
and Shanono (2019), Finite Element Analysis (FEA) is used
to perform several design studies under various restrictions
and the process entails breaking the framework down into
mesh-like smaller pieces. The fundamental idea underlying
the Finite Element Method is to combine several extremely
simple forms, such as triangles, in order to accurately
reproduce the original part, in place of any complicated
shapes with each one taking up a small but finite sub-domain
of the original component, the smaller, simpler forms are Figure 2. Sectional Free body diagram for the distributed
referred to as finite elements (Akin, 2010). case considered.
FA +FB =60331×dH = 12066 N
Methodology FA = 6033.1 N
The method of investigation using the analytical method FB =6033.1 N
ensured little or no assumptions. Shear force 𝑆𝐹 = 𝐹𝐴 − 𝐹𝑤 ∗×
Hydrostatic pressure on the sight-glass P=𝜌𝑔𝐻 At x=0, SF =FA= 6033.1 N
𝜌 = 𝑤𝑎𝑡𝑒𝑟 𝑑𝑒𝑛𝑠𝑖𝑡𝑦 = 1025 kg/m3 At x =𝐿/2 = 0.1; SF = FA-Fw*0.1 =0 N
g=gravity due to gravity At x=L = 0.2 ; SF = FA-Fw*0.2 = -6033.1 N
H= hydrostatic height over the glass =30m Maximum Shear stress= 𝜏 =
𝑀𝑎𝑥𝑖𝑚𝑢𝑚 𝑠ℎ𝑒𝑎𝑟 𝑓𝑜𝑟𝑐𝑒
=
P= 1025 x 9.81 x 30 = 301657.5 Pa 2
𝑆ℎ𝑒𝑎𝑟 𝑎𝑟𝑒𝑎
150,827.5 𝑁⁄𝑚

FUW Trends in Science & Technology Journal, www.ftstjournal.com

e-ISSN: 24085162; p-ISSN: 20485170; August, 2023: Vol. 8 No. 2 pp. 052 – 055 53
 Analytical and Numerical Methods in Structural Engineering Design Practice: Important Note for Young Engineers

Bending moment 𝐵𝑀 = 𝐹𝐴 ∗ 𝑥 − 𝐹𝑤 ∗ 𝑥 ∗ 𝑥/2 Figure 3. The free body representation of a concentrated load
At x=0; BM = 0 NM case
At x=𝑥/2 =0.1; BM = -301.6 NM Sum of vertical forces=0
At x=L=0.2; BM =0 NM FA + FB = F= 12066 N
The sectional free body diagram showing the sketch of the Take moment about point FA
SF and BM is shown in the result section as Figure 4. F*0.1=FB*0.2
Considering a concentrated load case: FB=6033 N
Force on the sight glass = P*A FA=6033 N
Assuming that the load will concentrate in the centroid of the Bending Moment BM;
sight glass: At point of FA;
Force F=12066.3 N BM =0 Nm
In a free body diagram, this is represented as in Figure 3 At point F;
BM=FA*0.1= 6033*0.1 = 603.3 Nm
At point FB;
BM=FA*0.2 – F*0.1 = 6033*0.2-12066*0.1 = 0 Nm
The stresses 𝜎, is computed for distributed and concentrated
load situation, and presented in Section 3.0.
This case is repeated on the Finite element method of
SolidWorks for various values of the smallest mesh sizes
possible and compared.

Results and Discussions

The application of methods described in 2.0 gives the
following results;

Figure 4: Sectional Free body diagrams showing the SF and BM for both distributed (left-hand-side) and
concentrated (right-hand-side) situations.
Given length of the acrylic material as L = 20cm and width If 𝑀′ = Maximum Bending Moment
W=20cm with thickness B =20mm 𝑀′ ∗ 𝑦
Distance 𝑦 from the centroid of the acrylic glass to the out- 𝐵𝑒𝑛𝑑𝑖𝑛𝑔 𝑠𝑡𝑟𝑒𝑠𝑠 𝜎1 =
𝐼
most stressed wall is approximated as half of the glass For the distributed situation, 𝜎1 =2.82 MPa
thickness as 0.01 m And for concentrated load situation, 𝜎1 =5.64 MPa
𝐿𝐵3 Shear stress= 𝜏 = 0.15𝑀𝑃𝑎 which is 2.7% of the
Second moment of area for this acrylic glass I= = 1.07
-6 4
12 concentrated load case or 5.3% of the distributed load case.
x10 m

Table 1: Comparing Analytical and Numerical FEM results of the Acrylic sight-glass of Wet Bell
Stress (FEM) -
Mesh element Numerical FEM Analytical stress Stress (Analytical)/ Stress(Numerical)
Load type size Stress (MPa) (MPa) FEM
Distributed 5mm 2.8 2.82 1.007143
Concentrated 5mm 5.2 5.64 1.084615

FUW Trends in Science & Technology Journal, www.ftstjournal.com

e-ISSN: 24085162; p-ISSN: 20485170; August, 2023: Vol. 8 No. 2 pp. 052 – 055 54
 Analytical and Numerical Methods in Structural Engineering Design Practice: Important Note for Young Engineers

The use of Finite Element Method of numerical solution solutions. The use of FoS in engineering practice to manage
using maximum mesh size of 5.0mm indicates no practical uncertainties is necessary for FEM users especially when
difference with analytical solutions. In fact, it can be noted reliability and safety of the structure is key.
that selection appropriate mesh sizes can converge to same
exact solution. This is very important for young engineers. References
Comparing this finding to the work of Giuseppe et al. (2019), Akin JE 2010. Finite Element Analysis Concepts via Solid
one can argue that their results could have been affected by Works. World Scientific Amzon.com Finite Element
any of the following: Analysis Concepts: Via Solidworks - John Edward Akin -
(1) The mesh size used in the FEM analysis. Google Books
(2) The analytical theory of Westergaard does not
agree completely with load on the structure, especially for Giraldo-Londoño O and Paulino, GH 2020. A unified
large load applications. approach for topology optimization with local stress
(3) Both the mesh size used and the Westergaard constraints considering various failure criteria: von Mises,
theory may not be adequately suitable to the applicable Drucker–Prager, Tresca, Mohr–Coulomb, Bresler–Pister
problem. and Willam–Warnke. Proceedings of the Royal Society
It is evidently clear from Table 2 that for smaller load A, 476(2238), 20190861.
application, the results of the analytical and numerical
computations tend to be converging. Giuseppe L, Paola DM, Laura M, Pablo Z 2019. Analytical
Table 2: Giuseppe et al. (2019) numerical FEM and and Numerical Approaches for Design of Stone Pavers in
Analytical results comparison. Urban Shared Areas. IOP Conf. Ser.: Mater. Sci. Eng.
Dimension
𝑆𝑡𝑟𝑒𝑠𝑠𝐴𝑛𝑎𝑙𝑦𝑡𝑢𝑐𝑎𝑙
pp.471 06203
of block Stress stress
Load(kN) (mxm) (FEM) (Analytical) 𝑆𝑡𝑟𝑒𝑠𝑠𝐹𝐸𝑀
Hearn EJ 1999. Mechanics Of Materials 2: An Introduction
50 0.42x.05 6.97 14.37 2.061693
To The Mechanics Of Elastic And Plastic Deformation Of
50 0.42x.08 4.64 7.19 1.549569 Solids And Structural Materials (3rd Ed.). University of
Warwick United Kingdom.
50 0.42x0.1 3.18 5.08 1.597484

50 0.42x0.12 2.76 3.8 1.376812 Karthik S, Tawfiq AM, Abdulaziz AM and Mohammed AD
50 0.42x0.15 1.9 2.65 1.394737
2017. Analytical and Numerical Design Analysis of
Concentric Tube Heat Exchangers – A Review . IOP Conf.
20 0.42x.05 4.61 7.85 1.70282 Ser.: Mater. Sci. Eng. 272 012006.
20 0.42x.08 2.68 3.7 1.380597
Kuang-Hua C. 2015. Chapter 17 - Design Optimization,
20 0.42x0.1 1.91 2.56 1.340314
Editor(s): Kuang-Hua Chang, e-Design, Academic Press,
20 0.42x0.12 1.4 1.89 1.35 Pages 907-1000, ISBN 9780123820389,
https://doi.org/10.1016/B978-0-12-382038-9.00017-X.
20 0.42x0.15 0.9 1.29 1.433333
(https://www.sciencedirect.com/science/article/pii/B978012
5 0.42x.05 1.17 2.13 1.820513 382038900017X)
5 0.42x.08 0.66 0.99 1.5
Muhammad A & Shanono HS 2021. Analysis of a Support
5 0.42x0.1 0.46 0.68 1.478261 Base Stand using Finite Element Analysis. Nigerian Journal
5 0.42x0.12 0.33 0.5 1.515152 of Technological Development, 18(1), pp.14-21
Muhammad A & Shanono HS 2019. Transient Analysis and
5 0.42x0.15 0.23 0.34 1.478261
Optimization of a Knuckle Joint. KINETIK, 4(2): pp.179-
18
Authors re-emphasis the importance of the use factor of
safety (FoS) in design and other engineering works. Though Nunez, MA, Brianço L & Dias D 2013. Analyses of a pile-
the thoughts of Stacey and Sharp (2007) was centered on UK supported embankment over soft clay: Full-scale
offshore industry, they insist of need to maintain high factor experiment, analytical and numerical approaches.
of safety (FoS) in Engineering. Stacey and Sharp (2007) Engineering Geology, 153 (8), pp.53-67.
presented minimum factor of safety of 2.5. It is noted that
this figure may be lower in non-offshore structures Stacey A & Sharp JV 2007. Safety factor requirements for
depending on the risk involved. The important consideration the offshore industry. Engineering Failure Analysis. 14(3),
here is that the application of the good safety factor in the pp.442-458
use of various FEM application should be seen as standard
practice for FEM users (see last columns of Tables 1 and 2).

Conclusion
It is important to accept that Finite Element of numerical
methods in various established software is a positive
development in engineering practice. The results, especially
when the model assumptions are realistic produce reliable

FUW Trends in Science & Technology Journal, www.ftstjournal.com

e-ISSN: 24085162; p-ISSN: 20485170; August, 2023: Vol. 8 No. 2 pp. 052 – 055 55