Menu

satavia

About satavia

Posts by Site Admin:

Junior and Senior DevOps Engineers

Junior and Senior DevOps Engineers

DevOps Engineers

Junior DevOps Engineer
Senior DevOps Engineer

SATAVIA is hiring DevOps Engineers to help bring about innovation of the aviation industry via data science and machine learning. We are helping to make decision making in aviation smarter and greener, using groundbreaking techniques.

You will be joining a multi-disciplinary team including engineers, mathematicians, climatologists, computer scientists and aerospace engineers who have large global scale datasets to analyse and global climate models running on a HPC environment. You will be helping us to present our datasets to our customers via API frameworks and innovative and engaging visualisations.

You’ll have an opportunity to work on our awesome technology stack including Azure, Linux, Python, PostgreSQL, Django, Docker and Dask. We run our own atmospheric models using High Performance Compute and serve data via a REST API. Teaser video shows us processing 4-dimensional climate data from our cloud Zarr store using Dask!

Click below for more details and to apply.

Junior DevOps Engineer
Senior DevOps Engineer
CorrosiveEnvironments

Corrosive Environments

The impact of corrosive environments on parked aircraft

SATAVIA tracks and quantifies corrosive environments to which parked aircraft are exposed at global airport locations. The number of aircraft parked for over 30 days at locations globally has rapidly increased over the past week, leading to corrosion and asset deterioration. In the animation below, aircraft fleets that have been parked for a minimum of 7 days on ground are shown in blue.

Aircraft fleets that have been parked for more than 30 days are shown in red (representing >90% of parked fleet at an airport), highlighting airports that have been dedicated to long-term aircraft storage. This means more maintenance work required to bring the aircraft back into service. DaaS (Data as a Service) is available via our secure API as part of our DecisionX product suite. Contact us for more information.

Figure 1
Figure 1
Figure 2
Figure 2

Joelle Garden MEng (Hons) MAPM

Joelle Garden MEng (Hons) MAPM

Joelle Garden MEng (Hons) MAPM

Head of Product and Delivery

Responsible for product management, leadership of key projects, development of business processes, and communication across departments and stakeholders.

Experienced in project management of aerospace engineering design and R&T projects in defence and commercial aviation sectors, at Marshall Aerospace and Defence Group and Ultra Electronics Precision Control Systems.

MEng in Aerospace Engineering from University of Bath, Member of the Association for Project Management.

“The unique blend of academic and industrial backgrounds within our team allows us to pivot to solve the biggest problems, and makes working at SATAVIA a thrilling ride!”

Fun fact: Had an unexpected close encounter with a white rhino while orienteering on a school trip, once upon a time.

Daniel Bannister BSc (Hons) MSc PhD

Daniel Bannister BSc (Hons) MSc PhD

Daniel Bannister BSc (Hons) MSc PhD 

Chief Scientific Officer

Daniel Bannister is an atmospheric scientist and climate modeller, responsible for driving and supporting the environmental science team.

Daniel holds a first-class Honours Degree in Geography from the University of Portsmouth and a Masters with Distinction in Atmospheric Science from the University of East Anglia. In 2015 he was awarded a PhD by the University of East Anglia and the British Antarctic Survey. Since completing his PhD, Daniel has worked on several projects on polar meteorology and climate change over complex terrain.

“The great thing about working at SATAVIA is that everyone supports each other! We all work together to build a genuinely collaborative environment.”

Parked aircraft

Parked aircraft due to COVID-19: Initial observations

Parked aircraft due to COVID-19: Initial observations

Figure 1 CFM56 engine flight numbers are decreasing globally, but regional flights within China are recovering.
Figure 1 CFM56 engine flight numbers are decreasing globally, but regional flights within China are recovering.

Problem statement

Flight restrictions and reduced passenger demand due to the ongoing Covid-19 pandemic has meant airlines around the world are grounding large numbers of aircraft. These aircraft are being left for a prolonged period subject to potentially damaging environmental conditions. Whilst grounding locations are often chosen to minimise damage, this is not always possible. Environmental damage such as corrosion sustained whilst grounded means the aircraft are likely to require additional maintenance before being brought back into service. Planning for this return to service will be a major problem for airlines and MROs, particularly with nearly two thirds of the global fleet currently grounded and much uncertainty around when and to what level service will return in different regions.

Solution statement

SATAVIA is tracking 20,000 active aircraft using terrestrial- and satellite-based ADS-B messages. Noisy data is cleaned giving precise numbers of flights with accurately inferred origin and destination information. Through this analysis we can construct a picture of global and regional flight numbers to help inform where service is resuming. We are also able to identify the location of grounded aircraft and couple this with our contaminant dataset to provide environmental risk assessments. As well as contaminant levels, corrosion risk factors are calculated to identify where aircraft are most at risk of sustaining damage whilst grounded.

MROs can use this information to prioritise maintenance requirements and help plan for required maintenance. Airlines could potentially reposition aircraft at high risk. Early notification of flights resuming could assist with scheduling the maintenance and re-certification required for return to service.

Figure 2 Corrosion risk at airports with grounded CFM56 aircraft
Figure 2 Corrosion risk at airports with grounded CFM56 aircraft

Case study

As a case study we tracked the global fleet of commercial aircraft using the CFM International CFM56 engines. Since the beginning of February, we have identified 22,000 engines in service, with 11,000 of those now grounded across 600 locations worldwide. Flight hour analysis shows significantly reduced activity but with some regions beginning to recover – in China for example. For the 600 airports at which aircraft are grounded we performed a corrosion risk assessment showing which aircraft are most at risk. This information could help inform maintenance schedules and planning for return to service.

Figure 3 Top 50 airports with grounded CFM56 aircraft and corrosion risk
Figure 3 Top 50 airports with grounded CFM56 aircraft and corrosion risk
FightAgainstCovid

Helping the fight against Covid-19

SATAVIA helps the fight against COVID-19

SATAVIA has been employing its data scientists’ expertise in the fight against COVID-19 as a supporter of the industrial alliance working on the UK government-led Ventilator Challenge.

Using in-house expertise, SATAVIA has developed a forecasting model for COVID-19 and has been using this to estimate ventilator requirements for the UK and other parts of the world.

Figure 1
Figure 1

The analysis predicts that the United States will enter into a period of peak ventilator demand over the coming days during which SATAVIA expects in-country COVID-19 mortalities to exceed 50,000.

The company is aiming to provide timely and accurate information on the pandemic to support the recovery of the aviation industry and enable airlines and airports to proactively prepare for the ‘return to service’ of the global fleet.

SATAVIA received support from Paul Perera at GKN Aerospace and Dr Ian McDonald at Microsoft through its involvement in the AI Factory – part of Microsoft for Startups.

While there is still much uncertainty around how the pandemic will evolve, this analysis provides a baseline for ventilator manufacturers and governments to assess demand, which will be generalised to other countries.

GroundedAircraft

Grounded flights

Grounded flights

More than 32% of the CFM56 engine fleet has been grounded since 25 February 2020. This engine type powers the A32X and B737 fleets – one of the most popular engines in service.

At SATAVIA we are tracking aircraft and engine flight hours, and where aircraft are being parked up. We analyse environmental conditions like humidity, rain, dust and air pollution while the aircraft is parked and provide insight to enable efficient return-to-service. 

Please contact us if you have any data requirements – we are here to assist.

Figure 1
Figure 1
Code

Autoscaling on Azure

Autoscaling on Azure

Here at SATAVIA we process many hundreds of thousands of flight trajectories in order to determine the exposure to various atmospheric contaminants.  Because we are a SaaS/DaaS(Software/Data as a Service) offering we are driven by customer requirements and timescales.  As such our service needs to be highly responsive to spikes in demand.  Our challenge is to deliver the best quality analysis to our customers, in an acceptable time scale while making the most efficient use of Azure compute resources.  We strive to have a highly responsive platform that scales up and down resources in response to demand automatically without manual intervention.

SATAVIA DecisionX’s trajectory exposure calculation relies on the existence of specialized memory optimized compute instances to complete trajectory exposure jobs. To achieve and maintain an acceptable job completion rate we often need to allocate multiple Azure D32 instances to the processing cluster. As this would drive the compute cost  up, we need to make sure that we utilize these resources as much as possible when processing takes place, but also that we have a way to automatically deallocate these compute resources when there are no jobs available. The problem in our case is that we cannot rely on traditional, simplistic resource utilization metrics (CPU, memory) to make the decision on when to allocate/deallocate resources. As a result, existing solutions such as Azure VM Scale Sets do not solve our problem as they scale based on CPU metrics.

Figure 1. Example view of our Airflow DecisionX platform autoscaling workflow. Airflow is able to create workflows of tasks that execute based on cron schedule
Figure 1. Example view of our Airflow DecisionX platform autoscaling workflow. Airflow is able to create workflows of tasks that execute based on cron schedule

Instead, we decided to use the number of jobs in a queue, as our scaling metric and we implemented a custom scaling solution based on this metric. We use Apache Airflow to create a workflow that runs in 10-minute periodic intervals. During every run the workflow queries a job table to find out how many jobs arrived in the queue and how many of them completed in the last 10 minutes. If that ratio is above a specific threshold then a decision is made to scale the number of instances allocated to the trajectory processing. However, instantiating fresh Azure VMs, means that we still need to account for provisioning and application deployment in an automated way once the VM is up and running.

The solution to this problem comes from our container orchestration and container cluster management platforms, Kubernetes and Rancher. Kubernetes is the industry standard orchestration and management platform for containerized applications deployed across clusters for VMs or bare metal servers. In addition, we use Rancher as a Kubernetes cluster management system, to easily provision and maintain different Kubernetes clusters (development and production grade). Rancher has connectors for all well-known cloud providers and can be used to provision managed clusters (e.g. AKS, EKS) or automatically deploy VMs against any cloud infrastructure and build the cluster from scratch. In addition, Rancher uses node templates to create groups of nodes that share similar characteristics (resources, network security rules etc.).

Figure 2. Rancher can organize Kubernetes nodes in groups, that share characteristics, such as VM size, network policies as well as custom labels.
Figure 2. Rancher can organize Kubernetes nodes in groups, that share characteristics, such as VM size, network policies as well as custom labels.

Once our Airflow workflow decides that the number of processing instances needs to scale, it sends a message to the Rancher REST API to scale the number of VMs assigned to a node group specifically created to host the trajectory exposure application. Subsequently, Rancher contacts the Azure API to create the VM, and once the VM is created Rancher automatically provisions the VM and attaches it to the existing Kubernetes cluster.

The new VM is now part of the K8S cluster, however we still need to deploy the application. Since the application responsible is containerised and deployed as a Kubernetes application, we had the option to use a special K8S controller called Deployment. A Kubernetes Deployment is responsible for parsing an application specification and maintain a number of pod replicas by reacting to changes and updates. However, by using its default scheduler Kubernetes could not guarantee the exact host that a pod would be deployed on. We needed exactly one pod per VM in the VM node group we created for processing trajectories.

Figure 3. Our application is deployed as a Kubernetes DaemonSet. The “nodeSelector” yaml tag will instruct the controller to deploy the application to nodes that are marked by the “worker=yes” label. A DaemonSet ensures that only one pod per node will be scheduled.

Autoscaling on Azure

Thus, we chose to switch to a Kubernetes DaemonSet. A DaemonSet is similar to a Deployment, however it deploys only one pod per node in the cluster. It can automatically scale the number of pods by reacting to changes in the number of available nodes in the cluster. The last piece of the puzzle that we needed to solve was to ensure that our trajectory exposure application would be scheduled only on that specific VM node group. Luckily Kubernetes offers the ability to narrow the number of schedulable nodes by specifying the .spec.template.spec.nodeSelector to the specification document of the DaemonSet. The node selector queries the Kubernetes API to match nodes that have a specific key value pair assigned to them called label.

Thus, by applying the appropriate labels in the VM node group, we were able to create a DaemonSet that deployed the application only to that node group. This also provided the benefit of reducing the complexity of our Airflow scripts. Airflow did not need to explicitly contact the Kubernetes API to increase the number of pods; once Rancher added an extra node to that specifically labeled node group, the Kubernetes DaemonSet would immediately detect the change and would schedule an new pod to the node. A single call to the Rancher API would create a new D32 instance, provision in, join it to the cluster and deploy our application instance, which in turn would immediately start processing jobs.

Figure 4. We allocate labels to Rancher node templates. Each node created against this template will have these Kubernetes labels.
Figure 4. We allocate labels to Rancher node templates. Each node created against this template will have these Kubernetes labels.

Our infrastructure can react to any number of jobs and fully scale to the maximum number of compute instances within 30 minutes. Similarly, it can as quickly scale down to the minimum once the number of jobs falls under a certain threshold. This enabled us to maintain high performance while keeping overall utilization and cost efficiency as close to 100% as possible.

Military aircraft landing

Business trajectories

Business trajectories

SATAVIA’s range of products solves different problems for different customers.

OEMs / MROs / Airlines
(Both military and civilian)

Common problems:

They want to out-perform competitors on engine, airframe and system reliability (predictive maintenance and future design), value for money (fuel efficiency and spares consumption) and environmental protection (carbon audit / credentials / offsets).

Our solutions:

SATAVIA delivers improved predictive maintenance, contributes to future aircraft design, increases fuel efficiency through engine wash and provides environmental management capabilities. Importantly, we are independent and auditable.

 

Lessors / Insurers / Financiers

Common problems:

They want to gain a competitive advantage by understanding the value of an engine, aircraft, or system. This involves developing a detailed knowledge of an aircraft’s (or fleet’s) state of corrosion, level of maintenance over time etc. This allows Lessors / Insurers / Financiers to more accurately calculate risk, which effects the need for cash reserves in their business and charging structures for clients.

Our solutions:

SATAVIA delivers uniquely detailed, auditable and independent historical (and predictive) business intelligence covering the corrosion of engines, aircraft and systems, plus locational data that provides insights on maintenance standards.

 

Case study – Major Middle Eastern airline

Through applying SATAVIA’s engine wash product, improved predictive maintenance and carbon offset audit capabilities the following savings can be achieved over one year.

Figure 1. Business case: Middle Eastern Airline Study.
Figure 1. Business case: Middle Eastern Airline Study.

Similar savings would be applicable to other airlines, or Military Fleets.

Img1

Contrail forecasting

Contrail forecasting case study

Background

SATAVIA can prevent aircraft contrails, eliminating up to 60% of aviation’s climate impact – that’s 2% of ALL human-induced climate impact!

Contrails are left behind by aircraft flying at high altitudes and are an everyday observable phenomenon. These “man-made” cirrus clouds can persist from seconds to hours, depending on the characteristics of temperature, humidity, wind, and the stability of the air aloft. Contrails that persist for several hours (and those that have significant horizontal spreading across the sky), are known to influence the radiation balance of the atmosphere, and, therefore, have potential global climate implications. In order to mitigate the overall impact of contrail-related clouds, understanding the aviation industry’s contribution to climate change is of vital importance.

Figure 1 – Contrail observations of LH430 (FRA-ORD) over city of Cambridge at approximately 1230 UTC, on April 10th 2020.
Figure 1 – Contrail observations of LH430 (FRA-ORD) over city of Cambridge at approximately 1230 UTC, on April 10th 2020.
Figure 2 – Forecasted air temperature profile (black) and critical air temperature for contrail formation (red) over Cambridgeshire on 10th April 2020 at 1300 UTC. The probability of persistent contrails is shown by the colours.
Figure 2 – Forecasted air temperature profile (black) and critical air temperature for contrail formation (red) over Cambridgeshire on 10th April 2020 at 1300 UTC. The probability of persistent contrails is shown by the colours.

Solution

SATAVIA has created a high-resolution (5km spatial resolution) regional climate model product to forecast the likelihood of contrail formation for flight corridors across Europe and Asia. We calculate the likelihood of contrail formation by comparing model-predicted temperature and humidity values to the critical values needed for their formation and maintenance. SATAVIA’s model could be used for short-term forecasts (up to 36 hours) of contrail layers. Contrail forecasts could be incorporated for flight planning purposes to intelligently re-route flights around those regions of the atmosphere that are susceptible to the effects of contrails.

Figure 3 – Terra satellite image over the British Isles (at approximately 11:41 UTC on April 10th 2020) showing a series of contrails over the Irish Sea
Figure 3 – Terra satellite image over the British Isles (at approximately 11:41 UTC on April 10th 2020) showing a series of contrails over the Irish Sea

Case study

On 10 April 2020, high pressure conditions over Western Europe led to calm, blue skies over much of the United Kingdom. At approximately 1230 UTC, a contrail was observed and photographed by three different observers (see Fig. 1), lasting longer than 5 minutes. The aircraft was identified as a Boeing 747-830, at an altitude of 36000 feet at the time of observation.

SATAVIA’s model predicted a layer of persistent contrails with high probability between 34000 and 40000 feet 1300 UTC (see Fig. 2). This is consistent with the visual observations at the time (Fig 1).

The nearest-in-time satellite imagery (Terra/MODIS) was approximately an hour earlier (at 11:41 UTC) than the photos shown in Fig.1 and clearly shows a series of persistent contrails left behind by several transatlantic flights over Wales and the Irish Sea (Fig 3).  SATAVIA’s contrail prediction system predicted an extensive area of potential for long-lasting contrails over most of the British Isles near this time, yet near zero chance over northern France.  This appears to match favourably with the satellite image.

Figure 4 – Terra/MODIS satellite image over the British Isles (left), and forecasted probability of persistent contrails at 36000 feet on 10th April 2020 (right)
Figure 4 – Terra/MODIS satellite image over the British Isles (left), and forecasted probability of persistent contrails at 36000 feet on 10th April 2020 (right)
Back To Top
©SATAVIA 2020. Company No: 8826207, The Registrar of Companies for England and Wales. VAT Reg No: 263199484. Registered address: Second Floor, Park House, Castle Park, Cambridge, CB3 0DU, UK. Satavia Privacy Notice