ExtendSim popularity in higher institutions has experienced phenomenal growth as researchers have learned of ExtendSim's ease of use combined with its high level of accuracy. ExtendSim has become an indispensable component supporting essential phases of innovative research projects in universities worldwide.
In response to the growing popularity of using ExtendSim in research projects, we established the ExtendSim Academic Research Grant program. Under the Grant, Imagine That Inc. supports students who are obtaining an advanced degree (Masters Thesis, PhD, or PostDoc) by subsidizing the cost of a full Model Developer Edition of ExtendSim for use by the student during the term of their research project. In exchange, the student provides a description of the research and quarterly updates throughout the term of the project. At the end of the research, all findings (ie. paper, project, etc.) and the ExtendSim model formulated are passed on to Imagine That Inc. for publication on the ExtendSim web site.
apply for an Academic Research Grant 
see completed Academic Research Grants
update form for current Academic Research Grant holders
Current Academic Research Grants
These research projects have been awarded an ExtendSim Academic Research Grant and are currently in progress. To learn about past projects completed under the ExtendSim Academic Research Grant program, please see Academic Research Grants Fulfilled.
Digital Twin • Critical Services Provider
Model-Based Systems Engineering
Xueping Li plus a team of Grad Students in the Department of Industrial & Systems Engineering
University of Tennessee, Knoxville
Grant awarded: May 10, 2022
To be completed: December 31, 2022
Hypothesis.
We propose the concept of using simulation as a soft digital twin to address complex decision-making processes in the age of digital transformation. The central hypothesis is that a well-developed simulation model is able to provide insights into critical business processes, conduct scenario analysis, and ultimately optimize the overall system performance.
Project description.
We adopt a multi-phased research plan for an undisclosed organization that provides critical systems.
Phase 1: Identify one area of interest (e.g., the maintenance organization) for detailed systems analysis. Identify model elements such as: variables, stocks, flows, and inter-relationships. Develop conceptual model structures. Establish functional requirements (e.g., real-time data acquisition, intelligent adaptation (machine learning), what-if scenarios, optimization capability, etc.) Define and establish the data sources, systems process mapping (including initial causal loop diagrams). Identify optimal modeling approaches (in consideration of multi-method modeling). Develop final causal flow diagrams, interfaces, and modeling elements for other appropriate methods.
Phase 2: Develop prototype model(s) for designated systems. Efforts include defining desired functionality, design and operational requirements, operating parameters, data resource needs, visualization, user interface, portability and access, system boundaries, and programing dictionaries (data, variables, parameters, links, interfaces, machine intelligence functionality etc.). Initial models will be constructed for proof-of-concept demonstration and preliminary V&V.
Phase 3: Complete design of the systems model, finalize verification and validation, complete UI and data visualization, complete documentation, and validate systems interface operability.
Questions this research will answer.
- RQ1: Will the soft digital twin, i.e., the simulation model, be able to produce results (e.g., throughput, overall job cycle time, resource utilization, etc.) that is comparable to historical system performance through a data-driven approach?
- RQ2: Will the soft digital twin help conduct what-if analysis with high confidence?
- RQ3: Will the soft digital twin help with system optimization under given certain operational constraints?
How research strategy will be designed to help answer these questions.
Please refer to the aforementioned multi-phase research plan.
How ExtendSim will be used in this project.
We plan to use ExtendSim to build the simulation model for the identified system (i.e., the maintenance system).
What is unique about this project?
The undisclosed organization that we work with provides critical services that demand high availability and any downtime will lead to exorbitant high costs.
Why the interest in the indicated subject?
Modeling and simulation have been one of my major research areas. The potential use of simulation as a digital twin is of particular interest to me.
What impact will this research make to the existing current state of knowledge?
We will have graduate research students working on the project. We intend to publish our research findings that will be disseminated widely.
Publications.
I have published 140+ peer-review journal and conference papers. Below are a few recent publications.
Zeyu Liu, Anahita Khojandi, Xueping Li, Akram Mohammed, Robert Davis, and Rishikesan Kamaleswarn (2022). A Machine Learning-Enabled Partially Observable Markov Decision Process Framework for Early Sepsis Prediction. INFORMS Journal on Computing.
Rodney Kizito, Zeyu Liu, Xueping Li, Kai Sun (2022). Multi-stage Stochastic Optimization Of Islanded Utility-microgrids After Natural Disasters. Operations Research Perspectives, No. 100235.
Chien-fei Chen, et al. (2022). Extreme Events, Energy Security and Equality through Micro and Macro Levels: Concepts, Challenges and Methods. Energy Research & Social Science, Vol. 85, No. 102401.
Yulin Sun, Cong Guo and Xueping Li (2022). An Order-Splitting Model for Supplier Selection and Order Allocation in a Multi-echelon Supply Chain. Computers and Operations Research, Vol. 137, No. 105515.
Rodney Kizito, Zeyu Liu, Xueping Li and Kai Sun (2021). Stochastic Optimization of Distributed Generator Location and Sizing in an Islanded Utility Microgrid During a Large-Scale Grid Disturbance. Sustainable Energy, Grids and Networks, Vol. 27, No. 100516.
Kaike Zhang, Xueping Li, and Mingzhou Jin (2021). Efficient solution methods for a general r-interdiction median problem with fortification. INFORMS Journal on Computing.
Rodney Kizito, Phillip Scruggs, Xueping Li, Michael DeVinney, Joseph Jansen, and Reid Kress (2021). Long Short-Term Memory Networks for Facility Infrastructure Failure and Remaining Useful Life Prediction. IEEE Access, Vol. 9, pp. 67585 - 67594. (doi.org/10.1109/ACCESS.2021.3077192).
Zeyu Liu, Anahita Khojandi, Akram Mohammed, Xueping Li, Lokesh K. Chinthala, Robert L. Davis, Rishikesan Kamaleswaran (2021). HeMA: A hierarchically enriched machine learning approach for managing false alarms in real time: A sepsis prediction case study. Computers in Biology and Medicine, Vol. 2021, No. 104255.
Chuang Liu, Huaping Chen, Xueping Li, Zeyu Liu (2021). A Scheduling Decision Support Model For Minimizing The Number Of Drones With Dynamic Package Arrivals And Personalized Deadlines. Expert Systems With Applications, Vol. 167, No. 114157.
Thomas Berg, Xueping Li, Tami Wyatt, Rupy Sawhney (2021). Agent-Based Modeling Simulation of Nurse Medication Administration Errors. Computers, and Nursing Informatics, Vol. 39, No. 4, pp. 187-197.
Yongzhen Li, Xueping Li, Jia Shu, Miao Song, Kaike Zhang (2021). A General Model and Efficient Algorithms for Reliable Facility Location Problem under Uncertain Disruptions. INFORMS Journal on Computing.
Updates.
Coming soon!
Digital Twin • Sow Farms
Anticipate future performance problems and issue alerts
Mihai Catalin Doja
Masters in Computer Science
University of Lleida, Spain
Grant awarded: May 17, 2023
To be completed: March 1, 2025
Hypothesis.
Digital twins can be useful to monitor farming operations at different degrees of digitalisation. They can anticipate future performance problems and issue alerts.
Project description.
The research plan will embrace three different tasks:
- Implement a simulation model of a sow farm.
- Develop metrics to measure technical and economic performance of the system.
- Integrate the model with an Excel spreadsheet to ease the use for practitioners.
Questions this research will answer.
The main answer is to test if a digital twin may help to delineate a path way for digitization in sow farms since many data sources could feed the model by using sensors. Not less important is to progress an interface development for an effective adoption of such decision support systems.
How research strategy will be designed to help answer these questions.
The support of her supervisor, Professor Lluis Miguel Pla Aragon, will be important in this strategy since he has already developed several approaches to the system Mihai that will have to review and update his existing models. For validation, they have an agreement with the Official Spanish databank of sow farms (i.e. BDPorc) who can provide them with real data to test and validate the model.
How ExtendSim will be used in this project.
The main use will be the development of the simulation model, using animation to refine the code, and connections to input/output with Excel spreadsheets to speed up the running of simulations allowing further and deeper data analysis.
What is unique about this project?
There are no simulation models for sow farms developed with the idea of digital twin (and digitization) in mind.
Why the interest in the indicated subject?
Because it is a strategic research line of the Department of Mathematics in the University of Lleida.
What impact will this research make to the existing current state of knowledge?
If the research progresses adequately, it is intended to go further in depth as Mihai is considering the development of a PhD.
Publications.
Mateo, J., Pagès-Bernaus, A., Pla-Aragones, L.M., Castells-Gasia, J.P., Babot-Gaspa, D. (2021) An Internet of Things Platform Based on Microservices and Cloud Paradigms for Livestock Sensors.
Pla-Aragones, L.M. 2021. The Evolution of DSS in the Pig Industry and Future Perspectives. In: Papathanasiou J., Zaraté P., Freire de Sousa J. (eds) EURO Working Group on DSS. Integrated Series in Information Systems. Springer, Cham.
Updates.
Coming soon!
Healthcare • Policy Setting for Operating Room Management
Surgery cancellation: causes, hospital-related strategies
Mona Koushan
Otago University
PhD in Management
Grant awarded: June 5, 2019
To be completed: May 31, 2021 & January 2022
Updated completion date: December 2022
Project description.
ExtendSim will be used to set a policy for OR management that balances the rate of surgery cancellation with hospital utilization taking into account the relationship and inter-relationship of different circumstances (e.g., scheduling policies, case mix, and hospital size) that affect policies.
Questions this research will answer.
- Which policy will be suitable for the hospital according to the circumstances?
- How does a manager set a policy that balances the rate of surgery cancellation and hospital utilization accounting for the circumstances they face?
How research strategy will be designed to help answer these questions.
To answer these questions, a discrete-event simulation will be used to investigate these relationships so Ms. Kaushan can develop a model/framework that will help the hospital manager to determine the most applicable policy considering the hospital’s circumstances.
How ExtendSim will be used in this project.
ExtendSim will be used to set a policy for OR management that balances the rate of surgery cancellation with hospital utilization taking into account the relationship and inter-relationship of different circumstances (e.g., scheduling policies, case mix, and hospital size) that affect policies.
What is unique about this project?
In general, studies consider a single policy to manage operation room capacity (and treat the policy as ‘fixed’ or ‘given’), but there are few articles that compare these policies and find which one can be more efficient (Y. B. Ferrand et al., 2014; Van Riet & Demeulemeester, 2015). While Duma & Aringhieri (2019) used each policy, they didn’t consider how hospital circumstances and various factors (outlined by Van Riet and Demeulemeester (2015), such as:
- the available number of ORs
- the available beds in downstream (ICU & PICU)
- operation duration characteristics
- scheduling policy
- patient volume
- case mix
...can affect the managers’ decision for setting an appropriate policy.
Why the interest in the indicated subject?
Increasing demand for healthcare service with limited resources has led to hospitals paying more attention to the use of resources. Given that the hospitals are naturally faced with different kinds of variability and uncertainty (such as surgery duration, length of stay (LOS), recovery duration and emergency arrivals), the fully planned of resources without any time buffers will require some level of surgery cancellations that will affect patient satisfaction. There are various factors that cause surgical cancellations. So, it become a main issue for the hospital manager to set a policy that balances the rate of surgery cancellation and hospital resource utilization, accounting for the hospital’s circumstances.
What impact will this research make to the existing current state of knowledge?
Today’s increasing demand for surgery and limited hospital resources has led to more surgery cancellations and resource under-utilization. Thus, solving this problem helps to even the patient flow while reducing the surgery cancellation (hospital-related causes) and managing the OR’s buffers with resource utilization according to the inherent uncertainty.
Publications.
Jahani, N., Jabalameli, M.S., Koushan, M., Rezaei Vandchali, H., Wood, L.C., "Bio-objective model for sustainable biofuel supply chain network design under uncertainties", Bioresource Technology Report (Derived from an external research project). Contribution: Validated the methodology and statistical analysis, interpreted, and wrote the final version of the manuscript. (Under review 2022).
Koushan, M., Wood, L.C., Greatbanks R. "A Scenario-based robust optimisation approach for multi-objective, multi-stage operating room scheduling with time and demand uncertainties", European Journal of Operational Research. (Under review 2022).
Koushan, M., Wood, L.C., Greatbanks R. Evaluating Factors Associated with the Cancellation and Delay of Elective Surgical Procedures: A Systematic Review, International Journal for Quality in Health Care. 2021, 33(2).
Duong, L. N. K.; Wang, J. X.; Wood, L. C.; Reiners, T.; Koushan, M. The Value of Incremental Environmental Sustainability Innovation in the Construction Industry: An Event Study. Constr. Manag. Econ. 2021, 0 (0), 1–21.
Koushan, M. Wood, L.C., Greatbanks, R. "Cause of surgery cancellation: A systematic literature review", Health Policy (submitted February 2020).
Koushan, M., Jolai, F., Wood, L.C., Hybrid Differential Evolution-Data mining (HDEDM) Algorithm for uncertain dynamic CMS problem, 48th International Conference of Computer and Industrial engineering, Auckland, New Zealand, December 2018.
Azadeh, A., Sheikhalishahi, M., Koushan, M. An integrated fuzzy DEA–Fuzzy simulation approach for optimization of operator allocation with learning effects in multi products CMS. Applied Mathematical Modelling (2013) 9922-9933.
Rafiei, H., Rabbani, M., Koushan, M., Effect of Motivation & Learning Curve In Dynamic Cell Formation And The Worker Assignment Problem, International Journal of Engineering Sciences & Research Technology (2012) 481-497.
Koushan, M., Jolai, F., Wood, L.C., Hybrid Differential Evolution-Data mining
(HDEDM) Algorithm for uncertain dynamic CMS problem, 48th International Conference of Computer and Industrial engineering, Auckland, New Zealand, December 2018.
Yadegari, M., Koushan, M., Inventory control, pricing of perishable items, taking into account time-dependent effects of inflation, 3th International Conference on Industrial Engineering and Sustainable Management, Isfahan, Iran, December 2016. (It was named one of the seven top articles).
Updates.
September 6, 2022 - I was granted my Ph.D. degree and the thesis is embargoed in the University library for further publication. After a discussion with my supervisor, I decided to add more analysis to make the paper ready for publishing in a leading journal.
The model is ready, only further analysis is required which might take up to the end of the year.
Small-sized hospital hybrid policyJanuary 24, 2021 - I am working on extracting articles from my thesis that require further analysis. I have submitted my PhD thesis at the end of December 2021, which the estimated time for oral examination would be in April 2022. After the oral examination, I need to apply examiners’ comments on my thesis. ExtendSim models are designed, but further analysis is required to present the output of the model. View screenshots of in-progress models here.
Extended Grant to September 2022 for Mona to improve the research.
September 21, 2021 - The models are designed for three policies separately and the experiments are designed. Now it is the time to run the model with different policies and scenarios. There are 90 different statuses that must be examined with these models. View screenshots of in-progress models here.
Extended Grant to January 2022 due to COVID related delays in data collection.
May 3, 2021 - Since this project is directly related to health bodies, my projects has delayed in data collection stage due to COVID-19, which has affected simulation study. Please extend the Grant through September 2021.
In this stage I am preparing input data for simulation model which are extracted from conducted surgery scheduling model with MATLAB. The ExtendSim model is almost designed and it is examined in the small-size of hospital. The model is going to be run with different scenarios and different size of hospitals. In this case some part of model might be changed.
July 27, 2020 - I am modeling my simulation model to check with a usual surgery scheduling how different policies can help hospitals to improve performance (less surgery cancellation and more resource utilization). I intend to check the model and different policies with comprehensive surgery scheduling modeling too. Comparing generated data from these two models, hospital managers can easily decide which policy works more efficiently given their hospital circumstances.
My model is almost complete. I just need to add some formulation to calculate rate of surgery cancellation rate and add inputs.
Optimization • Dairy Production Prediction
Use of AI tools and synergies in conjunction with operations research models
Osvaldo Palma Rubio
Doctorate of Mathematics
University of Lleida, Spain
Grant awarded: May 25, 2023
To be completed: September 1, 2026
Hypothesis.
The ability to predict cow's milk production has improved over the years with the use of data analytics tools. Many researchers mention the benefits of using simulation methods or artificial intelligence in the field of milk production. However, a combination of such techniques is not observed in the scientific literature, which presents an important research opportunity. Such a combination is expected to generate greater degrees of adjustment than by using the two techniques separately.
Project description.
- Year 1: Scoping review on dairy production prediction topics using artificial intelligence tools. Paper number 1.
- Year 2: Development of a discrete event simulation model that allows economic evaluation of productive policies and predictions of cow's milk production. Paper number 2.
- Year 3: Development of one or several neural networks for the prediction of cow's milk production and its comparison with traditional models. Paper number 3.
- Year 4: Development of a hybrid model that combines operational research methodologies and artificial intelligence for the productive improvement of dairy farms. Paper number 4.
Questions this research will answer.
How can the use of data analytics tools improve the prediction of cow's milk production?
How ExtendSim will be used in this project.
ExtendSim will be used to simulate the behavior of a dairy cattle and explore the net present value impact of different policies.
What is unique about this project?
According to a review of the current literature, there is no combination of simulation and artificial intelligence tools that try to make predictions about cow's milk production. The problem to be solved is important because the rapid growth of the world population will increase the demand for milk, therefore the use of technology will play an important role in being able to supply this demand.
Why the interest in the indicated subject?
The interest of this research is that it will be possible to study the production of cow's milk in order to meet the growing global demand for this important food through the use of data analysis tools.
What impact will this research make to the existing current state of knowledge?
Cow's milk production is important (Godfray et al., 2010). and depends on factors such as the productive potential of the animal and the conditions of exploitation. A common estimator of the current production of an animal is past production (time series) (Dongre et al., 2016), but there are other health or environmental factors that can condition the expression of this potential such as the presence of lameness and other diseases that can benefit from digital measurements of productive variables (Contla et al., 2021) such as body weight, diet control (composition and intake), early detection of pre- and postpartum diseases, assessment of body condition, etc. The main objective of the research is to explore the application of artificial intelligence (AI) tools for the analysis and prediction of productive variables such as cow's milk production. This study aims to be a contribution to data analytics (Data Analytics) combining traditional tools of operational research with AI.
Publications.
Currently in the last stage of preparation of my first publication consisting of the bibliographic review of this topic.
Updates.
Coming soon!
Production Flow • Mapping Bottlenecks
Rebecka Olsson
Lulea Technical University
Masters degree in Industrial Logistics
Grant awarded: February 14, 2023
To be completed: June 11, 2023
Hypothesis.
By simulating the provided, not yet existing production flow, an understanding of how the production flow will function can be determined. This understanding will hopefully help reduce and mitigate bottlenecks and other inefficacy in the production flow.
Project description.
Research begins with observation of an existing production flow at a company that maintains and services products and data collection of the current cycle times on each station in the production flow. The data will later be used to simulate the new production layout to determine:
- Will the layout will function.
- Where potential bottlenecks might occur and why.
This in combination with literature will help create the conclusion on how bottlenecks can be prevented in a new not yet existing production flow.
Questions this research will answer.
- How can simulation help identify and prevent bottlenecks in a new and not yet existing or tested production flow?
- How can bottlenecks be prevented and minimized in a new and not yet existing or tested production flow?
- What is the reason for the potential bottlenecks?
How research strategy will be designed to help answer these questions.
By simulating the production flow and in combination with relevant literature the simulation will show where in the production flow potential bottlenecks will occur and what is causing these bottlenecks. Regarding the result from the simulation in combination with relevant litterature, it will be possible to answer the question of how simulation can help identify and prevent bottlenecks.
What is unique about this project?
It is not often that a very complicated and expensive production flow is redesigned from scratch, which makes this an extraordinary opportunity to study which aspects should be taken into account in order to implement improvements against the previous production flow. Also the production line examined is not a manufacturing production but a production that maintains products. Hence, improvements are limited within a certain framework in that it is not possible to make changes in the product itself to make specific stations on the production line more efficient. The simulation is also done on a system that does not yet exist, which means that it is not possible to compare the result test in the simulation against reality, except against the old production line.
What impact will this research make to the existing current state of knowledge?
The research will highlight the impact and benefits simulation can have on the development and improvement of new production flows.
Updates.
March 31, 2023 - The first version of the model is almost complete. I have began with modeling the model as easy as possible andI plan to gradually add more advanced tools and parameter to make the simulation as realistic as possible. I have been working on the written report, data collection and modeling the simulation model. In the report, I have written an introduction, theory, method and empirical work. The data collection has involved observation of the production environment and collecting times for the simulation.
Resource Allocation • Multitasking in Livechat Support Centers
Santiago Gallino
University of Pennsylvania • The Wharton School
Academic Journal Publication
Grant awarded: January 12, 2023
To be completed: January 11, 2024
Project description.
When workers multitask, their performance depends on the multitasking intensity and duration.
Questions this research will answer.
- What are the effects of multitasking on chat duration?
- What are the effects of multi-branding on chat duration?
- What are the implications for optimal task assignment?
How research strategy will be designed to help answer these questions.
We have collected data from a large contact center and we plan to use ExtendSim to generalize the insights coming from this setting.
What is unique about this project?
Combining field data with a mathematical model and simulation outcomes uncovered this understudy effect.
What impact will this research make to the existing current state of knowledge?
Help better allocate resources in contact centers.
Publications.
Please see: https://oid.wharton.upenn.edu/profile/sgallino/
Updates.
Coming soon!
