Automated network design

The smart way to optimise infrastructure is to incorporate optimality into the design.”
Geoff Prince

In a recent presentation by Mike Quigley, NBN Co’s departing CEO, he mentioned that OSS/BSS gave him his biggest nightmares.

One of the NBN Co’s other significant problems he mentioned was the preparation of designs to service every home in Australia with broadband network services. Given the size of the challenge, NBN Co elicited the help of Biarri to develop software (called FOND — Fibre Optic Network Design) to automate network design for the Australia-wide fibre optic network layout.

It has been indicated that FOND has “reduced construction costs by approximately 20 per cent and manual planning time by 80 per cent.” It achieves this through the use of “mixed-integer programming, a branch of mathematics used by airlines and logistic companies to solve complex network problems in minutes.[which]… allows quick investigations (simulations) of how best to build, distribute and manage the fibre optic roll out.”

[The quotes above come from an article on the Australian Mathematical Sciences Institute website]

In a nutshell, FOND uses mathematical techniques to simulate many different design variations and choose the most optimal. Designing whole service areas manually is time-consuming enough, but then trying many different variations until you find the optimal one is not viable from a time and effort perspective.

I believe that you’ll begin to see more of these types of tools introduced into OSS software suites in coming years as they represent quantifiable and rapid returns on investment. For example, if the 20% construction cost reduction and 80% planning time reduction are accurate, then this represents many millions of dollars (perhaps billions) of savings on a $40B+ network like Australia’s NBN.

Not only that, but many organisations’s OSS suites already hold a majority of the raw data that tools like FOND need to perform their optimisation simulations.

The big caveat on these types of tools is the GIGO paradigm – Garbage In, Garbage Out. If the OSS data integrity is low or if the ducts listed in the OSS/GIS are irreparably damaged, then the application will be working from false assumptions.

Do you prepare many designs every year, have reliable OSS data and have repeatable designs / processes? This may make you a candidate for developing a viable business case to support an automated network design tool. FOND is not the only one already available on the market. I’d be happy to assist you with your business case and the process of choosing the best-fit for your business.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.