Using a Descriptive System Model to Scope Sizing Parameters

Using a Descriptive System Model to Scope Sizing Parameters

The author, as part of the Telecommunications Working Group in the International Council on Systems Engineering (INCOSE), has established a reference model for telecommunications service delivery. This model is used to discuss and demonstrate aspects of model-based systems engineering.
This presentation uses the reference model to demonstrate the derivation of network sizing parameters for an overall telecommunications system solution.
The approach taken has been to define sizing parameters for all elements, and connect these via constraint definitions. These relationships can be viewed through the Parametric and Constraint diagrams in SysML. Once these have been established, aspects of the overall service delivery (such as customer profile, geography) can be set, and constraints solved to size the various solution elements. Outcomes of this include calculations of not just the numbers of network elements needed, but also the sizing of maintenance and support systems, such as operational databases and customer service.
An important aspect of this approach is that the sizing constraint analysis is producing results directly using the MBSE system design model. As such, this can be translated directly into the project’s system requirements.

It has become common in the engineering of systems to classify models as “descriptive” and “analytical”. Using this distinction, the “descriptive” model describes the overall system-level architecture. As such, there should be one authoritative descriptive systems model for any project. Many analytical models can be used for areas of specific analysis, and these may or may not use the same software environment. Models may include simulations, and through time there is a need for these to move from purely constructive, into virtual and live simulation and analysis. Additionally, software environments are used to store and share agreed and endorsed requirements sets. There is an understood need for these to remain, in many cases separate from model-based systems engineering tools.

To successfully and consistently share information across these models, various approaches have been used, and projects must choose the appropriate method. Discussed in this paper are: direct linking of data (using interfaces such as the Open Standard for Lifecycle Collaboration – OSLC);batch data transfer, which will include linking and ability to apply later updates;and the execution of individual analysis models used to refine or change to the system definition.

Discussed in this presentation are the impact of these modelling and simulation techniques to the project, aspects of data management, the importance of identifying the specific purpose for tool use to meet a project need, and information security implications of choices made. Examples are shown where possible.

Speaker Bio

Daniel Spencer is a practicing Systems Engineer, specialising in Model-Based Systems Engineering (MBSE).

Mr Spencer founded Spencer Tech Pty Ltd in 2014 as a specialist provider of systems engineering and software services. Through Spencer Tech, he represents Zuken Vitech Inc. and Adelard LLC in Australia, New Zealand, and South-East Asia. Daniel has been responsible for product sales and technical support;development and delivery of training courses to clients;and specialist engineering consultancy services. He has also provided integration of digital engineering infrastructure, software development, and cloud-based IT infrastructure hosting.

An experienced Systems Engineer, he has worked on design and development of systems solutions for 20 years. Daniel holds a Bachelor of Engineering in Information Technology and Telecommunications with Honours from the University of Adelaide.

Daniel has contributed to the professional community at technical and management levels for the Systems Engineering Society of Australia (SESA), a technical society of Engineers Australia and the Australian chapter of the International Council on Systems Engineering (INCOSE).