Main Menu
Look beyond tech stacks and development frameworks
for remarkable business impact.

Build a product that is right for both the users and the business.

Our Work _ 
SABO Mobile

Shipping pump predictive maintenance pilot

Explore

Client

SABO Mobile
Industry
Manufacturing

Services

Data pipeline, Dashboards
We were brought onto an international consortium of companies to assist a German shipping pump manufacturer to develop and launch a scalable proof-of-concept model that eliminated the need to rebuild software once the solution is implemented on thousands of pumps. The goal of this model was for the manufacturer to help their clients transform their business from only selling pumps to providing equipment-as-a-service.

The Situation

Shipping pumps operate in complex environments inside large housings that are rarely visible from the outside. This makes it extremely hard to hear or see the typical warning signs or trends that predict an imminent failure or breakdown. This often leads to a reliance on reactive maintenance. But, maintaining equipment only after it breaks can mean unexpected downtime, emergencies, rush charges, overtime, and replacement of expensive parts.

One of the best pump reliability strategies is predictive maintenance. Effective predictive maintenance requires accurate data, and you have to keep in mind that the only reliable communication on ships is via satellite, which generally has a high cost and low bandwidth.

We were brought onto an international consortium of companies to assist a German shipping pump manufacturer to meeting two primary business goals: generate recurring revenue for their business and add continuous value for their clients.

The Objectives/goals

As part of the consortium, we helped to develop and launch a scalable proof-of-concept model that eliminated the need to rebuild software once the solution is implemented on thousands of pumps. The model needed to:

  • Deploy the sensors rapidly with context-appropriate connectivity.

  • Gather and securely store multiple sources of data in one place.

  • Present actionable insights to the right people at the right time.

  • Automate and optimise workflows and alarms to improve efficiency.

  • Predict events using artificial intelligence and machine learning.


Reaching these goals meant adding a predictive maintenance component to the manufacturer’s business model to potentially transform their client’s business from only selling pumps to providing equipment-as-a-service.

The Challenges

During the pilot programme, the risks and challenges became clear early and could be addressed and corrected. It’s worth spending a small amount of money and time to learn that predictive maintenance is viable and scalable.

  • The sensors used for predictive maintenance don’t last very long because shipping pumps operate in harsh environments (including high vibration, temperature changes, salt water, etc.).

  • Building a scalable proof-of-concept so that no software needs to be rebuilt when the solution is implemented on thousands of pumps.

  • Interpreting insights to take the right action at the right time.

  • There was no historical data to train a machine learning model.

The Solution

We spent a lot of time sourcing the right technology and hardware with the right characteristics:

1. Sensors that could withstand harsh marine environments.
2. The ability to process data offline on the ship and only sync when enough bandwidth is available.
3. Serverless architecture that could scale horizontally.
4. User-friendly data visualisation that enabled operators to take action.
5. Machine learning to predict machine breakdowns.

Key technologies used included: Off-the-shelf sensors, edge computing, Wifi and mobile satellite data link to the cloud, Amazon Web Services IoT Core, Grafana and email notifications.

The Results

The predictive maintenance model was employed to analyse and detect potential errors proactively. It used, for example, the machine’s vibrations and to avoid pump breakdowns and downtime. The ultimate goal is for this model never to stop learning. As it collects and processes increased data, it will become better trained to make better decisions.

We launched a successful pilot in less than 6 months and emphasised the benefit of initially spending a small amount of money first to establish whether the planned predictive maintenance model is viable and scalable before rolling it out.

Book your concept design and let’s start building better software, better!

Request for Access to Information

The following forms are available to download with regards to request for access to information:

REQUEST FOR ACCESS TO RECORD

OUTCOME OF REQUEST AND OF FEES PAYABLE

INTERNAL APPEAL FORM