IITB Logo Manish Kumar
Assistant Professor

Department of Civil Engineering
Indian Institute of Technology Bombay, Mumbai

Watch video of testing of fluid here

Try this web application for seismic isolation of bridge using elastomeric bearings
Seismic Isolation Web App

"Earthquakes don't kill people, structures do!"

Research Interests

Earthquake Engineering, Seismic Isolation, Energy Dissipation Devices, Performance-based analysis, Blast Engineering

I am a researcher working in the area of Structural and Earthquake Engineering. This website contains my professional bio and will guide you through my work. My research objectives include multi-hazard protection of critical infrastructure from extreme events such as earthquake, blast and impact. As part of my multi-displinary research efforts, I develop analytical models, perform numerical analysis, and conduct experiments. I have recently finished my Ph.D. dissertation work in which I investigated the feasility of seismic isolation technology to nuclear power plants using elastomeric bearings (Link). I have uploaded some of my research work to this website which might be useful to people with similar interest in Earthquake Engineering and Structural Dynamics.

PhD Research Area

Seismic Isolation of Nuclear Power Plants

Advisors: Dr. Andrew Whittaker
              Dr. Michael Constantinou

Supported By: Lawrence Berkeley National Laboratory
                      United States Nuclear Regulatory Commission


The nuclear accident at Fukushima Daiichi in March 2011 has led the nuclear community to consider the effects of beyond design basis loadings, including extreme earthquakes. Seismic isolation is being considered for new build nuclear power plant construction and design of isolation systems will have to consider these extreme loadings.

The United States Nuclear Regulatory Commission (USNRC) is sponsoring a research project that will quantify the response of elastomeric seismic isolation bearings under loadings associated with extreme earthquakes. Under design basis loadings, the shear strains in the elastomer will likely not exceed 150%, the bearings will not experience net axial tension, and the compression stiffness, which is important for the analysis of rocking in nuclear power plant structures (especially small modular reactors), will not vary substantially in the earthquake. The mechanical properties of the lead core in lead-rubber bearings may not change significantly. However, under extended or beyond design basis shaking, elastomer shear strains will likely exceed 300% in regions of high seismic hazard, bearings may experience net tension, the compression and tension stiffness will be affected by isolator lateral displacement, and the properties of the lead core will degrade due to substantial energy dissipation.

Phenomenological models, based on test data, are presented to describe the behavior of elastomeric isolation bearings in compression and tension, explicitly considering both the effects of lateral displacement and cyclic vertical and horizontal loading. The numerical models are coded in OpenSees and described. Results of numerical analysis are compared with test data to validate the numerical models.

Publications and Presentations

Google scholar profile:, Researchgate profile:

Journal Articles
1. Kumar, M., Whittaker, A., and Constantinou, M. (2015). "Experimental investigation of cavitation in elastomeric seismic isolation bearings." Engineering Structures (In press),
2. Kumar, M., Whittaker, A., and Constantinou, M. (2015). "Response of base-isolated nuclear structures to extreme earthquake shaking." Nuclear Engineering and Design,
3. Kumar, M., Whittaker, A., and Constantinou, M. (2014). "An advanced numerical model of elastomeric seismic isolation bearings." Earthquake Engineering & Structural Dynamics, 43(13), 1955-1974.
4.  Whittaker, A. S., Kumar, M., and Kumar, M. (2014). "Seismic isolation of nuclear power plants." Nuclear Engineering and Technology, 46(5), 569.

Conference Papers
1. Kumar, M., Whittaker, A., and Constantinou, M. (2015). "Verification and validation of models of elastomeric seismic isolation bearings." 23rd International Conference on Structural Mechanics in Reactor Technology (SMiRT 23), Manchester, UK.
2. Kumar, M., Whittaker, A., and Constantinou, M. (2013). "Mechanical properties of elastomeric seismic isolation bearings for analysis under extreme loadings." 22nd International Conference on Structural Mechanics in Reactor Technology (SMiRT 22), San Francisco, California.

Reports and Presentations
1. Kumar, M., Whittaker, A. S., and Constantinou, M. (2015). "Seismic isolation of nuclear power plants using elastomeric bearings (In press)." Report MCEER-15-0008, Multidisciplinary Center for Earthquake Engineering Research, State University of New York at Buffalo, NY.
2.  Kumar, M., Whittaker, A., and Constantinou, M. (2013). "Implementation of new mathematical models of elastomeric bearings for analysis under extreme loadings." Presented at the 2013 Conference of the ASCE Engineering Mechanics Institute, Illinois, USA.
3.  Kumar, M., Kumar, M., Whittaker, A., and Constantinou, M. (2013). "Advance numerical models for elastomeric & sliding isolators." Seismic-initiated events risk mitigation in lead-cooled reactors (SILER) international workshop, June 18-19 (presented by Dr. Andrew Whittaker), Rome, Italy.

Experimental data set
Kumar, M., Whittaker, A., and Constantinou, M. (2014). "Dataset for tension and shear testing of low damping rubber elastomeric bearings." SEESL, ed.University at Buffalo.

Software tools
Kumar, M. (2014). Computer Program ElastomericX, LeadRubberX, and HDR: User elements in OpenSees for analysis of elastomeric seismic isolation bearings under extreme loading, OpenSees, Buffalo, NY, (



A review of the modeling and analysis of seismic isolators in contemporary software programs

The numerical tools used by expert engineers to analyze seismically isolated structures at the time of this writing represent the state of practice. This section describes how elastomeric and sliding isolators are modeled in computer codes that are widely used in the United States, noting that numerical models of seismic isolation systems should a) include all isolators in the seismic isolation system, and b) the account for the spatial distribution of the isolators across the plan footprint of the isolated structure.

Link to the Report 

Performance-based analysis of a healthcare facility in California

A performance-based assessment is an integral part of the performance based design process. It expresses the amount of damage and the consequence of damage in terms of performance measures such as Probable Maximum Loss (PML), Scenario Expected Loss (SEL), and Scenario Upper Bound Loss (SUL), which are more relevant to the building owners and other stakeholders (FEMA, 2013). These performance measures are specified as building’s repair cost as a fraction of building’s total replacement cost. The consequence of damage to a building can be assessed through causalities, repair cost, repair time, and unsafe placarding. As it is difficult to predict the impact of damage precisely, a probabilistic framework is used to account for inherent uncertainties in the process of estimating the impact, and probable values of the damage and loss are calculated.

The methodology of seismic performance based assessment is explained here through an example of a hospital building located in California. The building was designed as per the existing seismic provisions in the building codes of 1960s. The improved knowledge on seismic behavior, hazard definition, and analysis methods provides the opportunity to assess the vulnerability of building to current seismic hazards. The seismic analysis of the building was finished as part of another course project (Advance Earthquake Engineering). Using the results obtained from the seismic response analysis, damage and loss calculations are performed. The Performance Assessment Calculation Tool developed as a part of ATC-58 project is used here to estimate the probable values of the damage and loss. PACT uses response data with fragilities of structural and non-structural components of the building to estimate the damage. Using the costs associated with replacement and repair of the building, it provides options available to decision makers. It is expected that this will help in making an informed decision on the continued operation of the building.

Link to the Presentation and Report* (under review, soon to be uploaded) 

Response of a steel column subject to blast loading

This section presents the analysis of a pin-ended 5 m tall W14x257 column of Grade A992 steel subjected to surface detonation using methods of UFC-3-340 and equivalent SDOF analysis. The blast loading history on the column was obtained from a hydrocode program, which provided a more accurate representation of the actual blast loading where the reflected pressure and impulse vary along the height of the column. A numerical model of the column was created in LSDyna and loading obtained from the hydrocode program was applied at five locations along the height of the column. Results of SDOF analysis were compared with LSDyna analysis.

Link to the Report

Surface burst modeling using Air3D

The blast analyses of structures using commercial finite element software programs, like LS-Dyna, model the blast loading as time-dependent point or distributed loads applied directly to the structure. These time-dependent loads can be determined using two methods: 1) Through the charts developed by Biggs (1964), DoD (2008) and PDC-TR-06-01 (2008), and 2) Through the computational fluid dynamics analysis. The first method was discussed in the previous report.

Surface blast loading on the column is obtained here using the software program Air3D, which uses computational fluid dynamics (CFD) approach. The time and space discretization used in Air3D provides a more realistic approach to simulate blast loading on structures. A three-dimensional analysis was performed here to generate the pressure histories and other blast loading parameters at monitoring points defined in the problem. The analysis results are verified against those reported in homework assignment 4. Analyses were performed using four different mesh sizes to check the sensitivity of response quantities to varying mesh sizes. Theory discussed in this report is based on the material presented in user’s manual of Air3D (Rose, 2006). The Air3D input file used here for the three-dimensional analysis was built on the example 3 in the user’s manual.

Link to the Report

Statistical estimation of soil parameters using CPT data

Uncertainties in soil properties are characterized using probabilistic analysis. Soil models are required for reliability analysis of soil at a site. Parameters of a soil model, which includes mean, variance, and correlation structure, are estimated using experimental data. The obtained soil models can be used for reliability analysis at another site that has similar geological properties. The soil properties are known to be spatially correlated. The finite-scale model is one of the two model types used for probabilistic modeling of soil (other being the fractal models), and is used here. The assumptions used for application of finite-scale models are investigated. Seven analysis cases are explored by implementing two algorithms to determine the soil parameters. The difference in results obtained using different analysis cases and algorithm are investigated.

Link to the Report

Finite element analysis of circular elastomeric bearings in compression

Elastomeric bearings are composite elements made up of natural or synthetic rubber layers bonded to reinforcing steel shims in alternate layers. Rubber, owing to its low shear modulus, accommodates large horizontal displacements, and steel shims combined with almost incompressible rubber provides high vertical stiffness. This behavior of elastomeric bearings is desired in isolation of civil engineering structures. Rubber bearings are used in varieties of applications including seismic isolation, bridge expansion bearings, vibration isolation etc. font size="2">

In analysis and design of elastomeric bearings, it is crucial to predict the compressive and shear stress and strains. Existing analytical solutions predict the desired quantities using “pressure solution” approach, which simplifies the complex problem to a relatively simple one using many assumptions. Although simplified, these solutions are in terms of infinite series and are not practical for use in design calculations. Recent work has been done to obtain simple analytical solution that can be used for practical calculations.

This work is proposed to model the composite elastomeric element using finite element method and assess the accuracy of simplified closed form solutions by comparing it with the results obtained from finite element analyses. A review of limitations and advantages of both solutions methods is to be presented.

Link to the Presentation and Report

Seismic Isolation of an existing bridge using elastomeric and lead rubber bearings

A bridge was selected to demonstrate the analysis and design of different types of seismic isolation systems. This project used elastomeric bearings at abutments and lead rubber bearings at the piers. Bridge was located in California subjected to seismic hazard determined by Caltrans ARS website. A simplified method was used to arrive at the preliminary design of elastomeric and lead rubber bearings using assumptions involving nominal properties, manufacturer’s guidelines and adequacy of selected bearings. Properties of preliminary bearing design were used to perform a Nonlinear Dynamic Analysis in SAP2000 and results of simplified analysis were compared with time history analysis. Seven of sets of ground motions were selected and scaled as per different criteria and a final scaling criterion was used for analysis in SAP2000. Adequacy analysis was performed for critical bearing. After little iteration, final sets of elastomeric and led rubber bearings were selected and performance of bridge with selected set of bearings was reported for both lower bound and upper bound analysis.

Link to the Report (50 pages)

Seismic retrofit of a six storey steel moment frame office building

An existing six storey steel moment frame building was found to have unacceptable seismic performance for a given seismic hazard. Building was required to be retrofitted. A nonlinear model of the bridge was created in RUAUMOKO for full nonlinear dynamic analysis using given ground motions. Project evaluated the seismic performance of existing building and then three different retrofit schemes were suggested as retrofit using a) Friction Dampers, b) Viscous Dampers and c) Base Isolation. A performance index, which was a function of interstorey drift, residual drift and peak floor acceleration, was suggested based on current literature available. Effectiveness of three retrofit strategies was compared with each other and with respect to the original building. Finally a retrofit strategy using base isolation and friction damper in first storey, which gave the best seismic performance among three, was proposed as final design retrofit strategy.

Download the Project Presentation
Link to the Report (150 pages)

Seismic assessment of a three span skewed steel girder bridge located in New Madrid seismic zone

PART1: An existing 3-span skewed steel girder bridge with wall pier was placed in New Madrid Seismic Zone characterized by its high seismicity. Bridge was design example 2 as one of the FHWA design examples prepared by Berger/ABAM engineers. Bridge was analyzed using different methods given in "AASHTO seismic provisions for design of bridges" and "MCEER ATC-49". Five different analysis methods included, Equivalent  SDOF analysis of the bridge, Single mode and multimodal response spectrum analysis, pushover analysis, capacity spectrum analysis, nonlinear time history analysis. Demands were obtained from different analysis methods and were checked against the capacity of the bridge. Expected performance of bridge with new seismic hazard was reported along with behavior of key elements. In the end methods to improve the performance of bridge was suggested.

Download the Project Presentation
Link to the Report (120 pages)

PART2: Peer review of similar work done by other group for a three span curved concrete box girder bridge characterized by a different seismic hazard.

Download the Project Presentation
Request for the full paper (60 pages)

Undergraduate Projects

Final Year B.Tech. Project: Determination of effective thickness of thin walled steel sections under compression 

The formula for equivalent thickness of multiple stiffened thin walled steel sections is reviewed. By carrying out the analysis of buckling strength of a thin walled section a more rational formula is suggested and comparison is done between  code formula and the suggested formula.

Link to the Full Project Report

B.Tech. Term Papers and Reports

1. Problem of "soft storey" in Earthquakes

Full Report

2. Dynamic analysis of suspension bridges 

Full Report

3. Seismic design of drawing hall building of IIT Kanpur

Full Report

4. Seismic assessment of a building near IIT Kanpur Campus gate

Full Report


Summer Internship 2007

Organization: Center for Offshore Foundation Systems, University of Western Australia

Location: Perth, Western Australia Department: Civil Engineering Duration: May 10th to July 15th, 2007

Background: Increasing exploration of new energy sources offshore in various regions of worlds has challenged the geotechnical engineers to come up with advance and economically efficient techniques to establish platforms and explore these sources offshore. Installation of suction caisson foundations to establish the offshore platform have been proved a very effective and cost effective method in this case.

Objective: My objective was to discuss the behavior the suction caisson installation in the case of layered soil by first deriving the formula for penetration resistance and predicting the behavior considering different cases of seepage of water in installation process.

My Role: I modeled the installation process schematically and derived the formula considering all different cases possible. I plotted the graph of penetration resistance and then explained the physical significance of the graph in reference to actual situation.

Challenges: There were very limited literature available on this topic, so I had to go through all the research papers published in the area so far to build the background of the topic. Topic was very new and there was no text available in the books as such, so I had to derive the formula only with the basic knowledge of Geotechnical Engineering after making appropriate and feasible assumptions.

Worked with: Dr. Christophe Gaudin, Manager, COFS(Centre for offshore foundation systems) University of Western Australia

Dr. Mark Cassidy, Director COFS(Centre for offshore foundation systems) University of Western Australia

Results: I was able to derive the formula for Penetration resistance of Suction caissons in layered soils considering all the different cases of seepage and explained the behavior of the installation process on the basis of graphs obtained.

Request for the paper

Advanced Coursework Done