Agribusiness Optimisation Solutions

Maths and Machine Learning for Agribusinesses

Mathematics powered by computers is changing the world we live in. At Biarri we see this everywhere, across every industry, and I’m sure you do too. Recently we have delivered a number of Machine Learning and Mathematical Optimisation solutions for Agriculture businesses in Australia and were fortunate enough to be invited to speak at the recent Case IH agri-business conference in Mackay.

Ash Nelson, Biarri’s co-founder, presented on Maths and Predictive Analytics for better business decisions. He described how our everyday lives are being changed by corporations leveraging large data sets, advanced statistical analysis and powerful computing resources. Ash then outlined how these same set of technologies can be utilised to improve business decisions in agriculture. This includes optimising agricultural supply chains and port operations, reducing unplanned equipment failures by using intelligent predictive maintenance algorithms or to improve health and safety outcomes for farm workers by better identifying areas of best practice to inform injury prevention initiatives.

Are you interested in leveraging your data using advanced maths to make better business decisions? Don’t hesitate to get in touch with our friendly team.

Healthcare Solutions Thumbnail

The Biarri Score: Predictive Analytics for better integrated healthcare

Many Health and Hospital services around the world have introduced integrated care activities to coordinate Primary (Community and GPs) and Secondary (Hospitals) care levels in order to increase the effectiveness of care being provided at all levels. 

Biarri delivered a predictive data analytics solution which successfully identified hospital patients in the community who were most likely to readmit to hospitals. Statistical analysis proved the Biarri Solution is able to identify the top 1% of the population at risk of hospitalisation with 90% accuracy.

In January 2018, Ron Calvert, the CEO of Gold Coast Health and Hospital Services (GCHHS) was in the news recognising Biarri’s predictive analytics in the form of a ‘Biarri Score’, which predicts an individual’s likelihood of readmission to hospital with ‘remarkable levels of accuracy’. Check out the article here.

The Biarri Score is now in place in the Gold Coast University Hospital emergency department. Also, patient criticality scores are integrated into local GP practices to provide intra-facility collaboration of healthcare services and prevent unnecessary admissions.

This application of Commercial Mathematics has resulted in increasingly better identification of patient who are likely to readmit, in order to improve the effectiveness of the Integrated Care activities and reduce demand on emergency departments.

 

Get in touch to learn more about Biarri’s Predictive Analytics capability or to learn more about applicaitons of Commercial Mathematics in other areas!

Biarri -Hospital and Health Services

Trimming away hospital costs: A bi-product of saving lives with statistics

Healthcare service delivery in most systems can be described as fragmented at best. In many healthcare systems, there has been very little continuity of care and integration for services provided by General Practices (GP), Hospital and Health Services (HHS) and other healthcare providers. Integrated Care is a worldwide trend in healthcare reforms that focuses on co-ordinating these different services.

Read more

Supply Chain, Logistics and Mining at BAM 2016

Supply Chain, Logistics and Mining at BAM 2016

The Biarri Applied Mathematics Conference is on again and registrations are open for 2016. This year BAM will be covering  a range of presentations across supply chain, logistics and mining optimisation. With our supply chains becoming more and more complex don’t miss out on exploring how optimisation can be used.

Extending the MIP toolbox to crack the Liner Shipping Fleet Repositioning Problem

Robin Pearce will be delving into the Liner Shipping Fleet Repositioning Problem (LSFRP) which involves repositioning ships between service routes while maximising profit. This presentation will demonstrate how this problem can become quite large, with multiple ships, thousands of potential cargo transfers and tens of thousands of arcs. A straightforward MIP implementation can solve small scale problems, however the problem quickly becomes intractable..

In this presentation Robin will show us how he has managed to reduce solve times from hours down to a few minutes.

Robin Pearce is a mathematics student at the University of Queensland. After studying a Bachelor of Science with Honours in Applied Mathematics at UQ, he spent two years as a Graduate Fellow with CSIRO working on three-dimensional microstructure modelling. He is now a PhD student with Michael Forbes, once again at UQ. His main topics of interest are the use of lazy constraints and disaggregated Benders decomposition for solving large and difficult integer and mixed-integer programs.

Optimal facility location and equipment selection for whey re-use

Whey is a by-product of cheese making that is a potentially important source of nutrients, but which currently goes to disposal in many parts of the world. In this presentation, Rasul Esmaeilbeigi will analyse the efficiency of investment in whey-processing with the aim of releasing the productive potential of currently unexploited whey supply chains. Rasul will describe a decision support model for production and distribution of products derived from whey that extends a globally inclusive facility location problem. The basic tenet of the model is that equipment selection during the initial stages of facility planning is critical, as capital costs in the early stages of supply chain design go into purchases of new machines and site conditioning. The model selects the optimal combination of whey processing equipment, facility locations and transportation routes subject to budget, equipment availability and final product requirements.

Rasul is currently a PhD. candidate in the school of mathematical and physical sciences at the University of Newcastle. he holds a master’s degree (2014) and a bachelor’s degree (2012) in Industrial Engineering. Rasul has expertise in the field of Mathematical Programming and Combinatorial Optimization and also general knowledge and experience of programming languages for solving large scale optimisation problems.

Multiple Yard Crane Scheduling with Variable Crane Handling Time and Uncertain Yard Truck Arrival Time

Container yard performance heavily depends on the efficient operations of yard cranes. Yong Wu will discuss the multiple yard crane scheduling problem with variable crane handling time and uncertain yard truck arrival time. Here the variable crane handling time refers to the variable time of handling each individual container, while the uncertain yard truck arrival time relates to the actual arrival time of trucks that are dispatched to either pick up or drop off containers. While there is a rich body of literature addresses the multiple yard crane scheduling problem in a deterministic operational context, there is a paucity of research incorporating these uncertain factors.

Dr Yong Wu is a Senior Lecturer at the Department of International Business and Asian Studies within the Griffith Business School. Yong holds a PhD in Operations Research and an MEng in Mechanical Engineering and has worked for The Logistics Institute – Asia Pacific, a joint venture between National University of Singapore and Georgia Institute of Technology (2005-2008), and the Institute for Logistics and Supply Chain Management, Victoria University, Australia (2008-2010). He teaches in the area of logistics and supply chain management and his research interests are in logistics and supply chain management, operations research and engineering optimisation.

Machine learning methods for mineral processing

Machine learning emerged as a subject area in the late 1950s; yet to date there has been little application of machine learning to mineral processing.

There are of course many ways that machine learning can be applied. Stephen Gay will pursue a probabilistic framework, strongly related to the new subbranch of mathematics called information theory.

The approach is to use far less samples than conventional methods and to infer many of the missing variables – indeed to infer the missing variables at a great level of depth (distribution of multimineral particles at each stream). By inferring this information we have a ‘snapshot’ of unit models for each series of plant data. Machine learning algorithms are then applied to parameterise the models according to operational parameters.

Dr. Stephen Gay originally graduated from University of Queensland [BSc (hons/Applied Maths)]. His domain areas have largely been in physical oceanography, mining (PhD), image analysis and geometric probability. The main area of mining is the development of software for optimising mineral processing plants. He received most of his grounding in mathematical modelling for mineral processing at the Julius Kruttschnitt Mineral Research Centre (JKMRC) – and in 2008 development his own independent consulting and contracting business which has since evolved into a startup Company: MIDAS Tech Intl. In 2014 he patented a method that enables the estimation of detailed mineral processing data from simple measurements – and has largely been focusing on getting interest in this new method from Mining Companies and Universities.

FTTx Planning and Design at BAM 2016

FTTx Planning & Design optimisation at BAM 2016

The Biarri Applied Mathematics Conference is on again and registrations are open for 2016. This year FTTx planning and design optimisation is a core component of the free 2-day conference so don’t miss out!

What are Fibre Optic Networks, how do we predict how expensive they will be and why do we do it?

Patrick Edwards will take BAM attendees through the complexities of a fibre network rollout and how it’s not always as straight forward as meets the eye. With billions of dollars being spent around the world on the deployment of these networks Patrick will explore the importance of optimisation and mathematics when predicting costs and different architectures.

Patrick Edwards has a background in mathematics, physics, programming and biochemistry. Patrick enjoys finding new ways to apply the skills from those areas to problems in the FTTx space. By conducting experiments for clients, Patrick helps companies across the telco industry make informed architectural and strategic decisions across their FTTx rollouts.

Why don’t Engineers and Mathematicians get along?

Alex Grime will be looking into the differences in how engineers and mathematicians think and speak across FTTx deployments and how that often gets in the way of successfully leveraging each other’s strengths:

  • Engineers want accurate, mathematicians want precise,
  • Engineers are interested in the destination, mathematicians are interested in the journey,
  • Engineers think 3 dimensionally, mathematicians think n dimensionally.

Alex Grime has over 20 years of experience in the telco industry across Network Strategy, Technology, Planning, Design, Cost Optimisation, and Operations. With a strong history in various roles across Optus, and NBN Alex is now one of the leading Telecommunications consultants for Biarri Networks.

How freedom to innovate is optimising global fibre rollouts.

Laura Smith will be discussing how FTTx networks are now being planned, designed and deployed with greater certainty, speed and at a lower cost by empowering smart mathematical minds. Through the use of optimisation, machine learning and other mathematical techniques, the entire industry is being re-imagined around us– and for the better.

Laura joined Biarri Networks after graduating with a Science degree in 2014. Starting as a member of the design team, she began taking on leadership roles and her focus changed to team development. Laura is passionate about process improvement and thrives on the challenges of working with a wide variety of people and clients.

The BAM Conference 2016

Registrations are now open and this year the conference will be held in Brisbane, Australia, on June 28 and 29 at QUT Gardens Point with support from The Queensland University of Technology, The Australian Mathematical Sciences Institute and Biarri.

Head over to the website to explore the other speakers, presenters and register now!

Biarri River to Rooftop Start

Biarri at Mater’s River to Rooftop 2015

Last week a bunch of the Biarri team gave up their Friday morning sleep-in to take on one of Brisbane’s tallest stair climbs to raise money and awareness for cancer.

The river to rooftop stair climb is an annual event held by the Mater foundation and is doing a fantastic job at raising money for prostate cancer research and finding a cure!

Biarri River to Rooftop Start

Tzara getting ready to climb!

Biarri River to Rooftop Tzara

Great views at the top, but exhausted

Biarri River to Rooftop Finished

Rav getting excited about the chance of winning a whipper snipper!

Biarri River to Rooftop Wippersniper

Good work to everyone that participated and we’re looking forward to a bigger and better stair climb next year!

About Mater Research

Mater Research is a world class institute which aims to discover, develop, translate, and commercialise medical research that can be translated into clinical care for the benefit of all. Mater Research is based in South Brisbane and specialises in cancer, maternity and obesity related research discovering ways to prevent and treat conditions affecting babies, children, adolescents and adults, helping them to lead healthy lives.

Supply Chain planning

Making the most of storage facilities in large construction projects

Large construction projects have a huge logistics component often requiring equipment from all over the globe to be brought together through a complex supply chain. In an ideal world, our suppliers would produce items on time and they would be transported to our construction contractors just as they needed them.

Unfortunately, we don’t work in an ideal world – suppliers can fail to produce on time, problems can occur in transit, and external forces outside our control can interrupt the construction schedule. For this reason, we want a contingency plan – usually in the form of a storage facility where we can hold spare or excess items if our supply and demand schedules don’t match up.

These facilities are used heavily over a short period of time, meaning that any inefficiencies quickly add up to large, unnecessary costs. However, given the size of these projects, identifying and fixing these issues is far from easy or intuitive. It is through analytics that you can drive quantitative answers and support effective decisions in your logistics and facility planning.

Read more

Mitigation by Iteration

Mitigation by Iteration – Facilitating the Optimisation Journey

For companies to remain competitive, they require smart systems that solve their unique day to day business problems. However, when applying these systems many decision makers get lost in the complexity due to limited communication and collaboration within the implementation process.

We are at the cutting edge of the latest optimisation methodologies and web technologies. However, unlike many other optimisation, and analytics companies out there, one of our main goals is to make powerful optimisation accessible in the real world to bring value to our clients.

We specialise in the development of web applications and smart optimisation engines– delivered in less time than you would probably think. It’s not unusual for us to go from an initial workshop with a client, to understanding their problem, and then having a fully functional optimiser in a production environment within three months.

On top of this the same people are often involved through the entire SDLC (software development life cycle) i.e. from the spec/design, theory, implementation and delivery/support. This reduces the overhead many organisations incur by having different people in business analytics and developer positions. The people implementing the solution actually understand and work with you to solve your problem.

Read more

Biarri Workbench - optimisaiton through the cloud

Optimisation through the cloud – The Biarri Workbench

Optimising your business; by operating in the most efficient and effective way; is essential in delivering a greater competitive advantage and is key to driving business success.

But, how can you drive innovation, empower your business decisions and survive in a highly competitive and volatile global marketplace?

A KPMG Report on elevating business in the cloud found that,

As cloud adoption picks up pace, cloud is poised not only to grow in scale, but will also increasingly impact more and more areas of the business. They do not only result in cost savings, but they can help organisations increase workforce flexibility, improve customer service, and enhance data analytics. In other words, the cloud should be considered a key enabler of the corporate strategy, driving strategic business transformations of all kinds.

Gartner Research supports this by naming the cloud as one of the top 10 strategic technology trends for 2015 that will have a significant impact on organisations during the next three years.

Agile Analytics driving business optimisaiton through the cloud

Anayltics delivered through the cloud empowers you to make adaptable decisions quickly, with far more rigour. Having access to your data anywhere, any time, on any device means that you no longer require large IT systems that are overly complex and not built for your specific problem. Cloud solutions allow you to scale up and down, and adapt depending on your specific target requirements. This means you can have point solutions targeting specific pain points within your business.

How does cloud based optimisation fit in my business?

Optimisaiton can be applied to most business problems. The question you should be asking yourself is; What is the best possible outcome of the decisions i’m about to make?

For your Supply ChainHow can I best support effective capital decisions to ensure end-to-end efficiency? – Learn More

For your Logistics – How should I best manage my fleet, workforce, facilities and work communications? – Learn More

For your WorkforceHow should I best plan my workforce across the next few hours, days, and months into the future? – Learn More

For your AnalyticsHow can I be sure that I am making the right decisions and considering all variables? – Learn More

How do we do it?

With our team of mathematicians, software developers and UI designers we use The Biarri Workbench which is an intuitive cloud based platform designed to support the rapid development on powerful web based software solutions.

With the powerful development platform of The Biarri Workbench, we are able to easily customise, alter, and build a solution for your specific business requirements.

The Workbench is Accessible. Empowering you to reduce your companies IT footprint and access world class optimisation anywhere, anytime, on any device.

The Workbench is Customisable. Built from the ground up to allow for rapid, bespoke deployment of software, built for your specific optimisation requirements.

The Workbench is Easy To Use. Through simple linear workflows, and customised visualisation widgets anyone in your business can easily master your software, reducing the need for long training and workforce upskilling.

The Workbench is Scalable. Regardless of business size or project complexity, through cloud based delivery, and bespoke software solutions built around your requirements, there is no more one size fits all approach.

The Workbench is Powerful. At the core of the Workbench are complex mathematical engines powered by industry grade commercial solvers. This means you can be certain in the justification around your decision making.

The Workbench is Efficient. Cloud based software delivery gives you the power to determine who sees what the data when– providing you with more control and rigour over your optimisation processes.

The Workbench is Secure. Security measures exceed both industry and customer requirements with the ability to easily accommodate your specific needs.

 Ask us how we can deliver optimisation via the cloud for your business!

Biarri FIFO Management

Grounding the complexity to Fly in Fly Out management

Being able to close the labour and skill gap is a critical factor in sustaining growth and maximising profitability for remote operations. It is imperative that companies have the tools and skills available to unravel the complexity to FIFO management.

FIFO workforces are commonly used by large infrastructure and resource projects in remote regions including rural and offshore. These regions often don’t have adequate infrastructure or an available local workforce with the right skillset which leads to companies requiring the use of workers from interstate and sometimes overseas.

The FIFO problem is complex for many companies. It involves determining efficient ways to move people via aircraft, taking into consideration: multiple projects at various phases over multiple locations, with a dynamic workforce utilising different skillsets on a variety of roster patterns, as well as using a fleet consisting of different types and numbers of aircraft.

Often the goal with FIFO management is to determine the number, and type, of aircraft needed in order to minimise cost whilst working with the opposing objectives of ensuring: the staff arrive before the start of their shift (but not too early), depart after the end of their shift (but not too late) and keeping travel durations to acceptable lengths (to ensure low fatigue).

Balancing FIFO Complexity

Analytics to break through the complexity

With this level of complexity, a traditional excel approach lacks the rigour and power to find the most efficient and effective results. As a result we’ve developed a number of different FIFO optimisers at Biarri to help ensure the best outcome for clients.

The reality is that there are often many more factors that need to be considered which complicates the problem further. Each FIFO optimisation problem often turns out to be quite different once the detail of the problem is better understood.

High Level FIFO Requirements

Some companies just want us to help them “define their fleet, or travel requirements” so they can then go out to tender (it also helps to keep the vendors honest), others actually want an operational tool. Others may be looking to see if there is a business case for upgrading an airport (e.g. if the airport is upgraded, then larger aircraft can be used which can reduce the need for bus in bus out (BIBO) which will alter their risk profile due to road km and can dramatically alter travel durations).

Specific FIFO requirements

Our clients often want different levels of detail in the solution. Some are happy with a solution that ensures adequate movements at the week level (e.g. 15 flights of aircraft type A between locations B and C per week), others want very detailed minute by minute schedules which take into account: turnaround time, time between takeoff and landing, number of aircraft gates with solutions showing exactly who is travelling on which flight and aircraft and when.

Across Multiple Projects

Our clients have also had multiple projects which are often on the go at the same time and sometimes different priorities are given to different projects. These priorities can be used to ensure that if all the people movement demands can’t be met, then the lower priority movements are less likely to be satisfied.

Optimising the time horizon

The optimisation time horizon can also vary significantly with some clients optimising over a 24 hour period (or even less if they want to re-optimise in the middle of the day due to unpredictable events such as delays due to weather) through to clients wanting higher level schedules over several years to help them make strategic decisions and determine how their fleet needs to change over time.

Understanding the constraints

Constraints such as: the maximum distance an aircraft can travel before needing to refuel, maintenance schedules and the refuelling locations themselves often also need to be considered. We’ve dealt with both fixed and rotary wing (helicopters) aircraft. Helicopters have the additional complication of sometimes having to take more fuel (and thus weight) to travel further, which results in the reduction of passengers because of the helicopter’s limited total payload capacity.

Finding the right FIFO parameters

We have outlined some of the parameters that our FIFO optimisers have considered. It is by no means comprehensive and we can always include new parameters if a different problem requires them but it gives a good understanding into the different variables that can, and should be considered.

Some of the typical inputs include:

Airport information

  • Location
  • Hours of operation
  • Refuelling capability
  • Refuelling duration
  • Availability (i.e. you can specify a start and end date for which the airport is available)

Aircraft information

  • Serial number
  • Category (e.g. fixed wing or rotary wing)
  • Type (e.g. DASH 8-200)
  • Average speed
  • Passenger seats
  • Maximum payload
  • Fuel density
  • Fuel tank capacity
  • Re-fuelling time
  • Fuel burn rate
  • Base location
  • Availability (i.e. you can specify a start and end date for which the aircraft is available)
  • Costs

Flight Legs

  • From location
  • To Location
  • Distance
  • Aircraft types able to fly this leg

Project priorities

People Movement Demands

  • Origin
  • Destination
  • Project
  • Number of passengers
  • From Date
  • To Date
  • Arrive Before (i.e. must arrive on their first working day of the roster by this time)
  • Depart After (i.e. must depart after this time on the last working day of the roster)
  • Roster Pattern (e.g. 14:14 = 14 days on, 14 days off)
  • Day of week (i.e. which day of the week can this person travel)
  • Group (demands can be grouped together to allow the user to specify which demands can be grouped on the same aircraft)

Some of the typical outputs include:

KPIs - There are around 40 KPIs, some of them are listed below

  • Total flights
  • Total distance flown
  • Total fuel burned
  • Total number of aircraft required
  • Utilisation Percentage
  • Total unused pax capacity
  • Total passenger demand
  • Total passenger demand satisfied

Resource Summary (i.e. which aircraft are required and when)

  • Serial number
  • Date
  • Total pax
  • Total hours flown
  • Total distance flown
  • Total fuel burned
  • Total flights
  • Total legs
  • Cost

Flight leg details (i.e. which flight legs are required and when)

  • Flight ID
  • Resource ID
  • Pax capacity
  • Available pax capacity (this is < pax capacity if the fuel weight is a limiting factor)
  • Total used pax
  • Utilisation Percentage
  • Departure location
  • Departure date and time
  • Arrival location
  • Arrival date and time
  • Day of week
  • Total distance
  • Total hours flown
  • Total fuel burned
  • Fuel weight at start of leg
  • Refuel at destination (true or false)
  • Turn around time
  • Cost

Flight leg pax details (i.e. which people movement demands travel on which flight legs)

  • Flight ID
  • Origin
  • Departure date and time
  • Destination
  • Arrival date and time
  • Project
  • Pax

Project summary (i.e. which demands from which projects were satisfied)

  • Project name
  • Total demand
  • Total satisfied demand
  • Total unsatisfied demand (e.g. this will be non zero if there is not enough capacity to transport demand)
  • Total impossible to satisfy demand (e.g. this will be non zero if a flight path has not been specified in the inputs that results in some demand being impossible to satisfy regardless of aircraft resources available)

Flight summary

  • Flight ID
  • Number of instances (i.e. how many times is this flight route flown at the same time – but on different dates)
  • Resource
  • Date of first flight
  • Date of last flight
  • Day of week
  • Departure time
  • Arrival time
  • Total people
  • Total distance
  • Total hours flown
  • Total fuel burned

Unravel the complexity to FIFO Management

The work we have done for companies such as Arrow, Origin, QGC, BMA, IBS, and Santos has shown us that despite having FIFO problems, they all required different approaches in order to achieve the right result.

This has demonstrated to us that when approaching a FIFO problem, where so many different variables have to be considered depending on the client, a standard approach (Commercial off the shelf product) and excel models will generally struggle with the complexity.

Having a tool built around specific variables demonstrates the benefits to bespoke solutions for FIFO problems.

Find out more about Biarri in Mining >>
Find out more about Biarri in Oil & Gas >>
Find out more about Biarri and FIFO Scheduling >>

Or, Get in contact so we can discuss your requirements.