Integration of COTS software products ARENA & CPLEX for an inventory/logistics problem

doi:10.1016/S0305-0548(03)00010-8

Computers & Operations Research

Volume 31, Issue 4, April 2004, Pages 533-547

https://doi.org/10.1016/S0305-0548(03)00010-8 Get rights and content

Abstract

This paper describes the integration of two commercially available, highly specialized, off-the-shelf software packages—CPLEX and ARENA. We present the need for such integration and the ways in which it differs from the more traditional methods of problem solving. The two products are described in detail followed by an example that requires this integration. A detailed account of the integration is also presented together with drawbacks of this approach.

Scope and purpose

The purpose of this paper is to demonstrate the mechanics of integrating two commonly used Operations Research software packages, CPLEX and ARENA. It then illustrates such integration by analyzing a sample supply chain application in the area of inventory/logistics. A discussion of the paper's findings and potential drawbacks to the approach are provided.

Section snippets

Motivation

Commercial-Off-The-Shelf (COTS) software products are products made available in the market to industries with specialized needs. These products are, in most cases, stand-alone applications. While the manufacturers of COTS products usually allow interfacing with programming languages (C, C++, BASIC, Java, etc.) and give their customers detailed instructions on how they can achieve this interface, they seldom document the possibility of integrating their COTS products with those of other

CPLEX and ARENA

In this section, we introduce the reader to the two COTS products in question, beginning with a brief introduction of the mathematical solver CPLEX and followed by a discussion on the simulation software ARENA.

A Sample supply chain problem requiring COTS integration

We look at a small segment of a Supply Chain with geographically dispersed distribution centers being replenished with product stock from the respective company's manufacturing facilities. These distribution centers are required to fulfill the requirements of customers as the demands for the various product-groups enter a central order management station. The function of the central order management station is to organize the distribution of these product groups to the customers from the

Integration

A major goal of this paper is to illustrate the integration of the COTS software CPLEX and ARENA. In this section, we provide the detailed steps needed to achieve this integration: All entities (demand arrivals) are made to pass through a VBA block as soon as they arrive into the system. As discussed earlier, the entry of an entity into a VBA block triggers a VBA_Block_Fire event. CPLEX is called from within the code for this event. Before CPLEX is called, however, certain attributes of the

Simulation experiment

The first point to be considered while designing a simulation experiment is deciding upon whether the system being modeled is best analyzed via a terminating simulation or a steady-state simulation. See Law and Kelton [14] for an extended discussion.

A terminating simulation is one that has a natural event (E) that specifies the length of each run. A non-terminating simulation, on the other hand, is one for which there is no natural event E to specify the length of a run. It is important to

Summary and limitations & alternatives to COTS integration

The COTS integration approach detailed in this paper is efficient in certain cases (e.g., the example discussed above). In this case the integration approach performs better than the traditional heuristic based approach (greedy order fulfillment policy) of Order Fulfillment.

There are, however, limitations to the use of COTS integration. As discussed in Nemhauser and Wosley [17], solution time for a MIP using the Branch-and-Bound scheme (such as that used by CPLEX) usually increases

Acknowledgements

This work was supported by a grant from Lockheed Martin Systems Integration—Owego. This support is gratefully acknowledged. The authors would also like to thank three anonymous referees for their helpful comments on earlier versions of this paper.

References (17)

M. Bogataj et al.
Combinatorial optimization of the location-inventory problem
International Journal of Production Economics
(1992)
A.K. Chakravarty et al.
Optimal multi-product inventory grouping for coordinated periodic replenishment under stochastic demand
Computers & Operations Research
(1988)
Yakimovich D, Bieman J, Basili VR. Software architecture classification for estimating the cost of COTS integration....
Stavridou V. COTS, integration and critical systems. COTS and safety critical systems IEEE colloquium (Digest) IEEE...
Verma D. Panel: commercial off-the-shelf (COTS) integration and support. Proceedings of the Annual Reliability and...
Stottlemyer AR, Hassett KM. Role of standards in COTS integration projects. Proceedings of the International...
ILOG Inc. ILOG CPLEX 6.5—user's manual. ILOG Inc., 1999....
ILOG Inc. ILOG CPLEX 6.5—reference manual. ILOG Inc., 1999....

There are more references available in the full text version of this article.

Cited by (30)

Modelling and simulation considerations for an end-to-end supply chain system
2020, Computers and Industrial Engineering
The efforts of this review paper are twofold: to provide an insightful examination of various contributions to knowledge surrounding simulation methods within an end-to-end supply chain and to guide research agenda by indicating generic elements required to model such systems using simulation. The authors examined 255 publications from 21 peer-reviewed journals in the field of an end-to-end supply chain and simulation using a systematic literature review approach. Each publication was thoroughly reviewed to capture best practices and key characteristics relative to simulation modelling techniques used in the context of complex end-to-end supply chain systems. This allowed for identification of generic elements required to model such systems, which were grouped into Structural, Computational and System Organization pillars. This research contributes to the body of knowledge by defining generic aspects of simulation modelling techniques used to study properties and attributes of complex end-to-end supply chains. The paper advances the theoretical understanding of the simulation methods used and applicability of simulation methodology in modelling end-to-end supply chain systems. The research presents the key findings from the use of simulation in modelling end-to-end supply chains and the main ways in which this modelling technique has informed research and practise.
Heuristics for Robots-Humans Tasks Assignment in a Containers Loading Center
2019, IFAC-PapersOnLine
To improve working conditions at sorting centers and to reduce the burden, La Poste wants to automate a part of the container-handling process. This mainly concerns containers that belong to the destinations with highest traffic since they demand an important effort and time amount from the operators especially in critical times like truck-departure times, when the operators must quickly load the containers in carts to be transported in trucks so that no delay will occur to the delivery date of the mail items. In this paper, we give a mixed integer linear program to assign the tasks to the robots, in order to maximize the workload of the robots and minimize the effort made by the operators. In addition, a greedy heuristic is proposed. Experiments performed on realistic data confirm the performance of the greedy algorithm. In addition, a sensitivity analysis of the heuristic is given to test its robustness on noisy data.
The application of discrete event simulation and system dynamics in the logistics and supply chain context
2012, Decision Support Systems
Discrete event simulation (DES) and system dynamics (SD) are two modelling approaches widely used as decision support tools in logistics and supply chain management (LSCM). A widely held belief exists that SD is mostly used to model problems at a strategic level, whereas DES is used at an operational/tactical level. This paper explores the application of DES and SD as decision support systems (DSS) for LSCM by looking at the nature and level of issues modelled. Peer reviewed journal papers that use these modelling approaches to study supply chains, published between 1996 and 2006 are reviewed. A total of 127 journal articles are analysed to identify the frequency with which the two simulation approaches are used as modelling tools for DSS in LSCM. Our findings suggest that DES has been used more frequently to model supply chains, with the exception of the bullwhip effect, which is mostly modelled using SD. Based on the most commonly used modelling approach, issues in LSCM are categorised into four groups: the DES domain, the SD domain, the common domain and the less common domain. The study furthermore suggests that in terms of the level of decision making involved, strategic or operational/tactical, there is no difference in the use of either DES or SD. The results of this study inform the existing literature about the use of DES and SD as DSS tools in LSCM.
Identification and fixing bottlenecks of a food manufacturing system using a simulation approach
2023, Scientific Reports
Identification and fixing bottlenecks of a food manufacturing system using a simulation approach
2023, Research Square
COTS software integration for simulation optimization coupling: case of ARENA and CPLEX products
2019, International Journal of Modelling and Simulation

View all citing articles on Scopus

View full text