JSON application programming interface for discrete event simulation data exchange

This research is conducted as part of a project that has the overall aim to develop an open source discrete event simulation (DES) platform that is expandable, and modular aiming to support the use of DES at multi-levels of manufacturing com- panies. The current work focuses on DES data exchange within this platform. The goal of this thesis is to develop a DES exchange interface between three di erent modules: (i) ManPy an open source discrete event simulation engine developed in Python on the SimPy library; (ii) A Knowledge Extraction (KE) tool used to populate the ManPy simulation engine from shop- oor data stored within an Enterprise Requirements Planning (ERP) or a Manufacturing Execution System (MES) to allow the potential for real-time simulation. The development of the tool is based on R scripting language, and di erent Python libraries; (iii) A Graphical User Interface (GUI) developed in JavaScript used to provide an interface in a similar manner to Commercial o -the-shelf (COTS) DES tools. In the literature review the main standards that could be used are reviewed. Based on this review and the requirements above, the data exchange format standard JavaScript Object Notation (JSON) was selected. The proposed solution accom- plishes interoperability between di erent modules using an open source, expand- able, and easy to adopt and maintain, in an all inclusive JSON le. In this way, the DES API model o ers the means for the con guration, synchronization, and maintenance of the di erent DES platform instances in a decentralized manner. The presented exchange data model o ers the means to represent (i) any results returned by the simulation engine, (ii) con guration data for various platform in- stances, and (iii) extra input data collected from various data sources. In addition, the DES API data model o ers an embedded JSON schema for the validation of the DES model graphs, and for the con guration of any GUI used for DES model design. A full description of the JSON DES API data model is provided with the di erent elements of the interface detailed. The DES API data model consists of six parts: (i) an element which is used for the con guration and the synchroniza- tion of the platform; (ii) an embedded JSON schema for the creation of consistent models; (iii) a graph that contains the topology of the model (iv) an element that contains DES related data, such as Bills Of Materials (BOM), or Work In Progress (WIP); (iv) an element that contains the results returned by the simu- lation engine; (v) and an element that contains general information crucial for a simulation run (e.g. maximum simulation run time, simulation time units). The proposed DES API data model is validated against three di erent real use case scenarios, placing emphasis on the various limitations they posed on the DES data exchange and on the API data model. The solution proposed for one pilot case is detailed in full with di erent limitations posed by the di erent speci ca- tions and pilot cases speci ed. Various expansions or enhancements that can be implemented are pointed out in the conclusions.