Speech by DPM and Minister for Trade and Industry Gan Kim Yong at the Launch of A*STAR's Sectoral AI Centre of Excellence for Manufacturing

Colleagues,

Friends,

Distinguished guests,

Ladies and Gentlemen,

Introduction

1. Good afternoon. It is my pleasure to join you all here today at the Grand Opening of the AI Centre of Excellence for Manufacturing (“AIMfg”).

Driving AI innovation within the manufacturing sector

2. I had announced our intention to set up the AIMfg six months ago in Parliament. The objective of the AIMfg was to convene stakeholders across industry, research and the start-up ecosystem to develop AI-enabled solutions that would address use cases in the manufacturing sector.

3. We believe that AI has the potential to transform our businesses. Not only in enhancing the efficiency of their processes to save costs and manpower, but also in creating intellectual property, new products or services, and unlock new opportunities in their respective businesses.

a. In this regard, we want to drive adoption and creation of AI-enabled services and solutions that will enable companies to reap the AI dividend.

4. The manufacturing sector is a key pillar of our economy, contributing about one-fifth of our GDP and providing employment to about one-eighth of our labour force.

5. We aim to grow the Value-Added for the manufacturing sector by 50%, over ten years, between 2020 to 2030, and to position Singapore as a global business, innovation, and talent hub for advanced manufacturing. However, our ability to do so will become increasingly challenged.

a. Domestically, we face tighter constraints of land, labour and carbon, which threaten to drive up the costs of doing business here. Globally, the competition is getting tougher, with many countries ramping up industrial policies to compete for new investments.

i. By coupling significant subsidies and loans with tariffs, export controls and domestic content rules, countries such as the US, South Korea and Japan aim to bolster their domestic manufacturing base, especially in key industries such as semiconductors.

b. Amidst a more constrained and competitive environment, we must double down on our strengths as an innovation-driven and productivity-driven economy.

Using AI to augment our manufacturing capabilities

6. The manufacturing sector can potentially benefit from AI, to overcome constraints and to remain competitive. To drive AI innovation within the manufacturing sector, we will adopt a two-pronged approach.

a. First, we will partner companies in Singapore to grow their AI capabilities and aim for them to set up Centres of Excellence (CoEs) in AI.

i. One example is the global biopharma company GSK. Through the Google AI Trailblazers programme, GSK developed an intelligent Good Manufacturing Practice assistant and an Investigation Report Maker for their Jurong manufacturing site that could save 450 man-days and 5,000 man-days per year respectively.

ii. GSK intends to continue developing this AI use case for deployment across its manufacturing site. We will continue to work with such like-minded partners to deepen their AI capabilities.

b. Second, we will set up sectoral-based CoEs in AI to build domain-specific capabilities.

i. The AIMfg is the first such sectoral-based CoE to build AI capabilities within our manufacturing sector.

7. By partnering companies in the manufacturing sector, the AIMfg will pool together meaningful use-cases and develop AI-driven solutions for these use-cases that can be shared across the sector.

a. For example, companies can make use of AI to enhance the productivity of manufacturing processes on the shopfloor. The solution can then be shared with other manufacturers to improve their productivity too.

i. One such use-case is in predictive maintenance. Today, maintenance is performed on machinery on a fixed schedule, to avoid cases of unplanned downtime that affect manufacturing operations. However, this may incur additional maintenance cost and potentially shorter equipment lifespan.

ii. With AI, companies can use data to optimise maintenance decisions based on actual machine health and perform maintenance only when necessary. This can potentially reduce –

I. Unplanned machine downtimes by up to 50%;

II. Cost of maintenance by 20 – 30%; and

III. Potentially extend the equipment life of machines by 10 – 15%.

IV. These improvements will enhance operational efficiency, increase productivity, and generate substantial cost savings. Once the solution is developed, it can be shared across manufacturing companies with similar maintenance needs.

b. Beyond the current set of use cases, we also hope to broaden the scope of potential use-cases to include areas like upstream product or component design.

i. There is potential to leverage Generative AI to enhance product design and development. With Generative AI, engineers can rapidly generate designs based on a set of parameters like structure, functions, materials, and cost. This will shorten the design cycle and improve responsiveness to market demand.

8. The AIMfg will employ an ecosystem approach to provide companies with access to key resources for AI innovation.

a. This includes access to the common AI models that are developed by AIMfg, and the sharing of best practices in processes like data management, model deployment, API management and AI governance.

b. The AIMfg will also partner with tech solution providers such as AWS, Microsoft and Siemens to support companies in accelerating their AI development and deployment efforts.

c. Besides models and computing resources, the AIMfg will also provide companies with common software tools and frameworks to facilitate collaboration and innovation.

d. Last not least, through technical consultancies, the AIMfg will guide them to deploy these solutions.

e. We are also developing an AI sandbox at the AIMfg.

i. The sandbox will allow more companies to experiment, test and explore AI-powered solutions, to discover the value of AI and hopefully invest in AI-driven innovation and solution.

9. AI talent is scarce, and the demand is rising. The AIMfg can also serve as a platform to crowd in talent from across the AI ecosystem and share the resources across the manufacturing sector.

a. For a start, the AIMfg will house a core-team of experts who are well versed with AI and its applications in the manufacturing domain.

i. These experts will come from A*STAR’s AI talent across the various A*STAR Research Institutes, such as the Institute of High Performance Computing and the Institute for Infocomm Research.

ii. They will also bring manufacturing domain expertise, from A*STAR’s Advanced Remanufacturing and Technology Centre (ARTC) and the Singapore Institute of Manufacturing Technology (SIMTech).

b. The AIMfg will also engage research talent from our Institutes of Higher Learning, as well as AI experts from tech solution providers, to complement A*STAR’s bench strength when needed in developing solutions for industry partners.

10. With a strong base of use-cases, talent and capability-building initiatives, the AIMfg can be a strategic enabler to drive the adoption of AI-enabled solutions within the manufacturing sector.

Conclusion

11. I would like to congratulate A*STAR on the launch of the AIMfg.

a. I am glad to hear that 13 companies have come on board the AIMfg. I am confident that the AIMfg will give them a strong start in their journey of AI transformation, so that they may be more productive, competitive and resilient. And I hope that more manufacturing companies will participate in the AIMfg.

b. This way, we can build a strong, innovative and vibrant manufacturing sector, that will enable us to achieve our Manufacturing 2030 vision, and continue to provide good jobs for Singaporeans.

12. Thank you.