Insights

Helping SMEs towards digitalisation

Small-scale manufacturing the smart way

‘Digital Manufacturing on a Shoestring’ is a major project led by a team at the IfM, and over the last year has been gathering pace to develop low-cost digital solutions for small to medium-sized enterprise (SME) manufacturers. Professor Duncan McFarlane, project lead and Head of the IfM’s Distributed Information and Automation Laboratory (DIAL), provides some updates on progress…

‘Digital Manufacturing on a Shoestring’ overview

Digital technologies promise exciting new possibilities in manufacturing. There are many opportunities for improving productivity and efficiency, both on the factory floor and in the back office. All of these improvements can help manufacturers operate more efficiently, drive down costs and strengthen customer relationships.

But there are major barriers for SME manufacturers looking to reap the benefits of digitalisation. Cost, complexity and the need for digital knowhow often make extensive digital transformation too difficult for small companies.

Most efforts in the digital manufacturing space have been focused on large scale solutions for big companies. Such solutions typically require significant initial investment and ongoing operating costs, as well as a need for digital skills within the organisation. In addition, they may require upgrades of industrial computing and communication environments to support advanced technological solutions. Cost and disruption are high. This situation risks leaving SMEs behind, and a different model is needed to support them as a vital part of the manufacturing sector.

‘Digital Manufacturing on a Shoestring’ is an ambitious project seeking to break down these barriers for SMEs. Funded by the Engineering and Physical Sciences Research Council (EPSRC), our research team at the IfM is collaborating with researchers at the University of Nottingham and with SME manufacturers.

SMEs are looking for inexpensive and easy digital manufacturing solutions to specific problems. They typically haven’t got large specialised IT departments. So, as well as being low-cost, solutions need to be easy to use. Our aim is to enable these firms to harness the benefits of digitalisation without the excessive risk and cost of implementing systems designed for large companies.

The Shoestring project is investigating how existing and readily available digital technologies can be implemented on a low-cost basis to support growth and productivity in SMEs.

The Shoestring approach

  • Helping SMEs implement digital solutions to improve productivity
  • Solutions must be accessible, pragmatic, useful, affordable
  • Based on low-cost components
  • Accounting for regulations, safety, security
  • Engaging students and local IT community
  • Solutions are repeatable, reusable, integrable

We are working with a large number of SME manufacturers to identify requirements, and create a catalogue of customisable digital solutions, complying with regulatory, safety and security requirements.

The solutions will be based on commercially available technologies for mobile computing, sensing and AI, identifying how these can be exploited, and tackling the issues associated with integrating these safely and securely into a small scale manufacturing environment.

The project involves partners such as Raspberry Pi and Siemens, along with teams of developers. It also brings together researchers in industrial information and control systems, machining and automation, asset management and maintenance, and draws on interdisciplinary expertise from computer science, economics and data analytics. In direct partnership with industry, we are working closely with manufacturing networks including the Scottish Manufacturing Advisory Service (SMAS), Make UK and the Nottingham Manufacturing Network, with involvement from SME partner companies from the start.

Our approach

The project has several phases, starting with understanding the needs and priorities of SMEs, before moving on to development, testing and pilots.

Phase 1: Digital requirements assessment

In the first project phase, we are spending time listening to SME manufacturers, to understand their needs and to identify common requirements between companies.

This is happening through requirements gathering workshops around the UK and in other countries. In the workshops, SMEs are being asked to discuss their priority business areas for digitalisation and provide scores to rank their operational requirements. We are also conducting onsite visits to partner SMEs to evaluate digitalisation ‘readiness’ and needs.

From these activities, the research team is classifying SME challenges and priorities, and creating a list of top priority requirements for proposed low-cost digital solutions.

Phase 2: Solutions development

Requirements gathered in phase 1 are being fed into development of low-cost digital options. This stage involves designing, modelling and developing architectures for low-cost component integration and infrastructure options.

The idea is to create ‘building blocks’ of existing low-cost technologies which can be connected into integrated digital architectures. Solutions incorporate consumer-grade components (such as Raspberry Pis) and low-cost sensors (such as bluetooth low energy beacons, off-the-shelf sensors, OS sensors and motion cameras).

The toolbox components can be combined with existing cloud computing platforms, human-machine interaction (consumer-grade AR/VR technologies), IIoT suites and interfaces such as iPads and Alexa (Amazon Echo) to create standard combinations which can form adaptable solutions.

Phase 3: Prototyping and pilot testing

The Shoestring team will be building proof-of-concept demonstrators both in research labs and in SME partner operations. A comprehensive testing and validation programme will demonstrate evidence of effectiveness of proposed digital solutions. Initial testing and evaluation, using SME industrial scenarios, will be done in labs at Cambridge and Nottingham. This will involve statistical assessments of prototype operations. Following this, there will be pilot studies onsite at partner SMEs, with the creation of case studies of early solutions.

Phase 4: Incremental integration

The team will identify how solutions can be implemented and integrated successfully within manufacturing environments. They will be looking at how solutions can be added one by one – in an incremental manner. This is to minimise infrastructure investment up front, and to maximise the ability for new solutions to make full use of services already provided in existing ones.

Connecting solutions is integral to the project’s approach. While it may seem cost effective to solve problems one at a time, the challenge is to avoid creating isolated, unconnectable solutions which won’t integrate with future developments. The Shoestring team is seeking to develop frameworks which allow for future integration.

The project team plans to work with BSI and others to support different types of standards across data, communications, services and architectures. They are also accounting for security and compatibility.

Phase 5: Engagement and dissemination

During the project, there will be open engagement with as many SMEs as possible, and firms are actively encouraged to get involved. We will be running regular open workshops – the first of these took place in Cambridge on 24th September and was well-attended by SMEs, providers and manufacturing networks.

A key outcome from the project is to create a ‘library’ of solutions which can be accessed widely by SMEs, as well as sharing lessons learned from exploring low-cost digitalisation. The team plans to develop a demonstration platform, to provide examples and visibility for addressing specific challenges within individual SMEs.

Digital solutions development: Engaging IT SMEs, students and hackathons

Fitting the entire ‘Shoestring’ approach, the project is actively engaging with IT solutions SMEs in the development processes, including workshops and events involving IT developers and start-ups.

Student participation is also strongly encouraged, with hackathons for engineering and computer science students, competitions for low-cost digital manufacturing solutions, and potentially sponsorship of relevant student projects.

Examples of Shoestring solution development

Working from the priority list of solutions from the SME requirements gathering workshops, the team is already busy building solutions. Here are a few examples.

Part detection and tracking

Dr Lavindra de Silva

This solution uses QR codes to detect and keep track of parts in a factory. When parts are received from suppliers, a QR code is attached to each part and they are placed inside boxes or trays with an ID. The boxes are placed flat on a tote, which the operator then manually ‘scans’ under a webcam or low cost camera that is connected to a Raspberry Pi and screen. The system will then identify all of the parts inside the box, scanning all of the QR codes simultaneously, and store selected information about each of the parts. The information can then be searched later, so the contents of the box can be checked. New parts can be scanned individually and matched with boxes containing similar parts.

Augmented reality status viewer

Dr Greg Hawkridge

Enabling a factory supervisor to view the status of the shop-floor systems, such as inventory management, this solution uses augmented reality on a smart phone/tablet. It can overlay status information from machines, such as stock levels, on a tablet or mobile device when the screen is held up to look at the machine. The idea is that a worker or manager can be walking down the shop floor and look at a particular machine through the app on their mobile device, and get an instant status check.

Machine monitoring

Dr Yedige Tlegenov

This solution provides a low-cost machine monitoring system that can be attached to machines, such as a 3D printer or a CNC machine, using sensors attached to Raspberry Pis and open source software libraries. It can obtain the status (such as temperature, through put, run time, etc) of the machine. Data from the sensors is sent to an open source cloud-based platform to provide monitoring information, including alerts of machine errors, and can be viewed real time on any device.

Voice assisted assembly

Dr Greg Hawkridge

This solution uses a digital voice assistant to help a worker complete an assembly task. The worker can request particular fasteners (nuts and screws) from an Alexa-enabled device which processes verbal commands. This is integrated with a turntable via a programmable logical controller. The system will provide them with the correct item and give them verbal feedback saying what the fastener or part is that’s been provided. This makes workers jobs easier and reduces errors.

See these demos in action in a series of videos here:

www.digitalshoestring.net/how/demos-and-examples/

 

To join the ‘Shoestring’ movement…

If you would like to be involved in this project, register your interest by emailing dial-admin@eng.cam.ac.uk   

Stay up-to-date on the progress of this project, including dates and times of workshops and demonstrations by signing up to alerts through the website: www.digitalshoestring.net