Page 135 - IJOCTA-15-2

An International Journal of Optimization and Control: Theories & Applications
ISSN: 2146-0957 eISSN: 2146-5703
Vol.15, No.2, pp.330-342 (2025)
https://doi.org/10.36922/ijocta.1694

RESEARCH ARTICLE

End of day process optimization through multi-mode resource
constrained project scheduling - A banking case study

1*
1
Ezgi Karabulut T¨urkseven , Ezgi Gen¸c , and Ilgın S¸afak 2
1
Department of Industrial Engineering, Sabanci University, Istanbul, Turkey
2
Fibabanka R&D Center, Istanbul, Turkey
ezgi.turkseven@sabanciuniv.edu, ezgi.genc@alumni.sabanciuniv.edu, Ilgin.safak@fibabanka.com.tr
ARTICLE INFO ABSTRACT

Article History: An End-of-Day process is a batch job that includes a sequence of programs,
Received: September 29, 2024 wherein tasks are completed automatically at times specified by a scheduler.
Accepted: February 8, 2025 The efficient allocation of resources for the timely execution of tasks allows
Published Online: April 8, 2025 a company to reduce the overall time needed for the completion of the work
Keywords: and improve customer satisfaction by delivering orders on time. This paper
Multi-mode resource constrained presents a case study of a Turkish bank with the objective of minimizing the
project scheduling duration of the end-of-day process through the optimization of work scheduling
Thread allocation and resource allocation. This problem is modeled as a multi-mode resource-
Simulated annealing constrained project scheduling problem, optimally resolved by mixed-integer
End of Day process optimization programming, and approximated via simulated annealing heuristic. The sched-
Banking case study uler in this paper can also be used to assess the importance of scheduling and
crashing tasks, along with the sufficiency of the infrastructure to optimize the
AMS Classification:
End-of-Day process.
90C11; 90C27

1. Introduction is required to complete each task within a speci-
fied period, commonly referred to as the process-
ing time.
Financial institutions such as banks carry out a Efficient allocation of resources to perform
financial End of Day (EOD) process to wrap up tasks on time will allow a company to achieve its
the day’s operations and prepare the system for objectives, such as minimizing the time required
the next day’s transactions. This is considered to to complete all tasks, maximizing customer satis-
be one of the most important parts of the mainte- faction, maximizing plant throughput, maximiz-
1
nance of any banking application. The EOD pro- ing profit, or minimizing production costs. Sev-
cess is employed by banks to produce and dissem- eral methods can be used to optimize the EOD
inate reports to stakeholders for regulatory over- process, including code optimization, code par-
sight, business users, and downstream systems. allelization, data partitioning, and load balanc-
An EOD process is a batch job including a se- ing. However, deploying applications in the same
quence of programs, wherein tasks are automat- physical location and using identical technology
ically executed at specified intervals via a sched- significantly increases the probability of achiev-
uler. Batch processing is a technique for handling ing a coordinated EOD process for each applica-
substantial volumes of data and transmitting this tion, leading to a considerably faster information
data in groups to the designated system. Sched- flow and a reduced duration of the EOD process.
uling is the process of allocating limited resources In large-scale enterprise applications that are in-
to process tasks over some period in an optimal herently dispersed and utilize diverse technolo-
manner. A certain amount of specified resources gies, such as microservices, the coordination of an
*Corresponding Author
330