A transparent UK-led supply chain: British Textile Week 2022
We caught up with Keric Morris from IBM as part of British Textile Week, to discuss a project we’ve been working on together improve transparency of fashion supply chains that could create a new benchmark for the UK industry.
Keric Morris, the executive partner for enterprise strategy and the global energy and sustainability lead at IBM Global Business Services, explained some of the benefits for UK textile companies and opportunities to get involved with the project.
The Sustainable Supply Chain Optimisation (SSCO) project is a Research and Innovation Initiative jointly funded by the UK Government through UK Research and Innovation and Innovate UK and the project partners are UKFT, Tech Data, the Future Fashion Factory and IBM. SSCO is a nine-month pilot project which started in April 2021 which is aiming to improve visibility and transparency throughout the fashion supply chain. The pilot organisations working with the project were Next, New Look, Laxtons and N brown and H&M (the COS brand).
Can you give us an overview of the project and how IBM got to be involved?
IBM has been working within this space, looking at transparency and sustainability from a supply chain perspective for a while, starting off more in the food industry than in fashion.
The start was really around making sure that product recalls could be done effectively, and understanding where traceability was from that sort of perspective, but we were having conversations two years ago around how we could apply that to fashion.
We found that while there were lots of good components which were being built within the industry, there wasn’t sort of a ubiquitous kind of platform, which could work on the technology side. We then went to have a chat with government about how we could take this forward.
The project is all about changing the traceability piece from being a product backwards to a fibre forwards. What is generally happening at the moment is using elements of the product backwards to understand the tiers of organisations that are involved. But what we’re doing is taking data in directly from the suppliers – production data- and then we can aggregate it. For example, when an invoice is then created, it effectively connects all the dots back up, backwards all the way through the supply chain. We’ve been working with a variety of different organisations such as those mentioned above but also the Better Cotton Initiative (BCI), the Higg Index and Textiles Exchange to look at how not only does this gives you a kind of a golden thread of transparency, but how you can then layer in additional data. How can you connect that data so that for BCI, for example, you can connect all the way back to the farm? And making sure that what we’re creating is a kind of a truly holistic process. The idea with this is also that that the process is then open to any sort of retailer and any supplier. Once you’ve collected the data, because you’re collecting effectively all the production data for that supplier, if other retailers have been using that same supplier, then they can also harvest that data too. So what you’re creating is sort of a cross-industry platform, which anybody can then tap into as part of the process moving forward.
What we have built through the pilot is an industry grade, scalable transparency solution that can ingest as much data as there is out there for as many tiers as we like. We can layer in additional data. So if you want to put water certificates or carbon elements, we can layer all of that in there. We can then have the API to then interrogate that data, to do the analysis to extract the data and then to present it. Working on the ingestion of data hasn’t been straightforward as we have had to use both voice and text ingestion. So we’ve been working with farmers in Turkey, for example, to test those solutions out. We’re effectively ingesting photographs and documents, then using AI to interpret that data and ingest that. So it’s not that some of these pieces of data are actually particularly technically simple, but there are solutions for it.
From a technical standpoint, we’ve proven the concept with cotton for both certified and non-certified products. The main issue is really around some of the changes that we need to then make as an industry. The biggest piece here is how a retailer can work with suppliers to make this as simple as possible for them. Supply chains are set up in a way that you can create competitive advantages for the various different stages so sharing production data is not normally done and not normally asked for.
There is a whole stakeholder discussion that you have to have around why we are doing this. What data is required? How do we sort of kind of get that data sort of repeatedly? Is there need for things like audit? Some of the main challenges are around engaging with stakeholders and the processes.
What we also need to get better at I think, as an industry, is forcing the issue to say we are asking for this data and we need it as part of the process. But it’s not built into the way that we do things. So we need to change our processes and build in supply chain transparency to cover everything from forced labour and minimum wage, water to carbon. These all need to be brought in as part and parcel of what we do as a core process, because you can’t really sort of do these as an adjunct later on. So it’s actually some of those more process, people, stakeholder issues, which are probably the, the trickier part of the puzzle.
But again, what we’re looking to try and do is from a technical standpoint is to let them make that as simple as possible. What you’re then looking to do is to make data collection part of the process and then it be provided in as simple and as repeatable way as possible. We have spent a lot of time on understanding those elements of the process. So that you can even do it quickly, you can do it once, or you can change your process. If I’m working at the farm level, what we’re actually saying is we need your data so when you’re handing over your bales of cotton to the ginner, we also need you to take a quick photograph at that point and send it to this system. And you do that every single time so it becomes part of the process. There are some process components that we do need to change as part of this as well and it’s those elements, knitting those pieces together, which is probably the most complex.
Why start with cotton?
We were looking for was a process that would be as useful as possible to the retailers we’re working with, and all of them had had cotton as part of their mix. The second reason was actually because of its complexity, which sounds like a weird thing to do. But because it’s a multi-tiered process and not an industrialised process like polyester, it meant you had to get to farm level. We wanted to understand all of the complexities that potentially would crop up. So, how do I get a farming collective in India to ingest data rather than a large corporate who produces polyester? From a data perspective, it was looking to systems of record and extracting data through chat or through voice. So we wanted it to be as complex as possible. The final reason was that it already has a certified (organic) and uncertified route so we could test those elements out. By certified we were talking about BCI or the GOTs certified route. Those work differently because for GOTs the certificate actually goes with the fabric as it moves, while BCI is more of a mass balanced approach. We wanted to understand those types of issues and what you then need to do in order to be able to get the data to kind of flow effectively across them. We wanted to make sure that as we were working on the research pilot, which we completed at the end of end of last year, that we could actually sort of get into as many of the problems and the complexities as we possibly could as part of that process.
How could the process be applied to other fibres?
From a process perspective, they’re pretty similar. I’m not saying that their supply chains are similar, but the process that we are going to do is very similar. So most of what we’re doing is a kind of a rinse and repeat, so what you’re actually doing is not far removed as you move from one fibre to another. It’s just really understanding what’s the core of that supply chain. How does it work? You need to know what the key steps are and who are the key players. You also need to understand their level of sophistication, by which I mean technical sophistication. Do they have systems of record for example? How can you leverage that? You also want to have a look at the certification components so is the product certified or not? Beyond that, really what you’re doing is the same thing. We’re asking suppliers to provide us with production data as part of what they do in their daily work. We want to understand when the batch came in, what was the invoice number and the same when the batch count went out. What was the invoice number? What’s the transformation event, which has happened in between? And how has that happened? And then we’re off to the next fibre. So it doesn’t matter whether they’re a polyester supplier, a wool supplier, a cotton supplier. It’s the same core process.
How did you work with a UK textile company like Laxtons during the pilot?
Laxtons were interesting because they didn’t have cotton as part of their raw material mix and they are in a different place within the process as a supplier rather than a retailer. They would have to provide the data rather than trying to access the data at the retail end. What they were very keen to do is really understand how they could get some better data into what they do. They wanted to prove where their wool comes from and the properties associated with that particular wool. There is value in providing that to customers that they supply. We’re seeing a greater move to transparency, not just from a reporting perspective, but people are looking at where a jumper is from, how you know where it is from and what the is nature of that particular product.
How far has it been transported? How are the sheep being treated as part of that? That’s starting to become more and more of value. So we were looking at them from that side of the process and it was a relatively easy place to start.
The issue with the cotton component is that the suppliers we’re talking to are in Turkey or Bangladesh. We couldn’t just go and visit them quickly and it was COVID too. It meant you couldn’t physically get your hands on the process. So the opportunity to have a look at it from a wool perspective, in the UK, working with people, like the British Wool Marketing Board, allowed us to have that more ‘hands on’ side of things. We did go and talk to farmers, and we have had conversations through the British Wool Marketing Board as well. And, what we’ve sort of done from that perspective is the same thing we did with cotton; map the process and understood what the issues are which are associated with it. What we haven’t done yet is take it to the next level, because of the economics of wool and the value of the fleece, particularly going in into the process in the UK at the moment. The traceability in the same way as we’re doing for the retailers doesn’t have the same level of value. But what we’re doing now is how to use the same sort of techniques, to support people like the Wool Board to then get a greater degree of transparency for the product through the supply chain from that perspective, because the way that works, and the fact that they are an aggregator pulling together a variety of different farmers’ products, there is a natural break in the process. We want to look how you can undo that break. So that’s progress in a slightly different route. We’re not really looking at it from a customer perspective, but rather we’re looking at it from a supply chain side of things. That hasn’t progressed as far as cotton has, but it’s still an ongoing conversation.
The pilot partners were generally large retailers as they have some of the most complex supply chains. But the aim is to create a system that can be used by businesses of all sizes, and all parts of the supply chain. Why is that important and how would it work?
It’s very important because what we are seeing is a great proliferation in all elements on the supply side, and actually on the retail side, of organisations that are looking to take new positions, create products, and do things in ways which are somewhat different to kind of the classic retailers. And they form a very large part of the industry, both from a volume perspective, and from the number of organisations. What we’ve always been looking at is how you can create something, which is not too expensive to access for those players. The way the platform is set up is that you know the costs for being on it and the access that you gain effectively can then be run on a service basis. So there aren’t massive costs to get on board there. Then what we’re then doing is, depending on the volume throughput, you can then effectively pay for what you get as part of that process. The whole point is how we can get the cost down as low as possible and how it can be as repeatable as possible. And that’s why we’re sort of kind of systemising, but also sharing. One of the things that we’re starting to kind of look at for the moment is, can we get a group of smaller suppliers together, to run a pilot with them now that we’ve done the bigger organisations so that we can then start getting into some of the issues which are associated with those smaller organisations.
A big question is who has to pay for this. Is this done at the retail end on a per garment basis? Or do you make the suppliers pay? What we’re trying to do is to make it so that and the barriers to entry are as small as possible to everybody within that and to create an equitable split across the play. There are also issues around if I am an early adopter, am I paying all this money for the development? That’s where UK government has come in, because they’ve paid for quite the development and all of the partners have also paid for quite a lot of that development already. Hopefully now we’ve overcome a lot of the challenges and now we need to get this into an economical and repeatable process.
The good thing is that because of the way it’s set up, when a supplier and the data is there, you shouldn’t have an onboarding cost. Once there are a good set of suppliers on there, the cost for someone joining the platform, if their suppliers already there is very small, because they’re already there. But then it’s question of how do you then equitably split across all of the people are on board, because if I’m an early adopter, then none of my suppliers will be there, because I’m one of the first. If I’m coming on later then a lot of my suppliers are going to be on there. So my onboarding costs are going to be less, but actually, should the person who was first pay for me? There are still some conversations to be had around that economics piece, and the cross subsidisation components. And how you make sure sure it is accessible to smaller players, and not prohibitive in costs. From the technology side of things, it isn’t actually that expensive. And funnily enough, you think there’s lots of data in this and there is but in the grand scheme of things, it’s not that much. That’s in comparison to other industries that we work in. So the volume of throughputs, and the technical back end that’s required to manage, it’s actually relatively containable so those costs aren’t huge. I think the economics of this are playing quite nicely in our favour, but we still need to do some work around that.
There is demand at retail level for this but how would suppliers not linked to those retailers get involved? Some of UK, textile companies, for example, might want to come on without their retailers yet as a way to show they are proactively embracing transparency.
We’ve already had some of those conversations as well. As we were going through the process, the Bangladeshi High Commission got in touch to say they were presenting to the Bangladesh fashion ministry, and they are really interested in being part of this process, because they want to be able to prove the efficacy of what they’re doing and be part of the transparency process. So, to your point, we’re getting suppliers saying we want to be part of this. It’s not then just about who is currently in your supply chain, it potentially is a way for others to then say you may not use us as suppliers right now but actually we are good. We can prove we were really good, and we’d like to be part of the process. So we want to be on here, because it’s not just about who you’re buying from today, it’s also who you might want to buy from tomorrow. It’s not massively complicated to do it, because the process is the same, irrespective of whether you are a supplier or retailer and irrespective of material. Once you start putting more and more materials on, or if you blend materials that’s relatively simple too because it’s just two data streams blending, then we can manage all of those things, because of the way the system has been built.
How could a project like this benefit the UK supply chain, in particular, UK textile companies, as it progresses?
What we’re looking to do, assuming that people want to buy into the process, is make this a UK standard. Because we believe, and I think we’ve proven through the pilot components of what we’re now doing and starting to roll it out, that we can do this effectively at scale and repeatedly. What we could then have as an industry is traceability and being able to prove that traceability; showing the steps that we’re being able to take in order to drive the sustainability components. And actually not just sustainability. One of the other things around supply chain transparency is that it also gives you is the ability to do better supply chain management. So if we know there are outages in certain parts of the world, there are shipping issues in a port, or there are container issues which we’ve been having lots of, and they’re impacting a particular region, this tool will be able to tell you who you have got in another region because we geoposition everyone. It could tell you how that might affect your supply chain.
So this isn’t just about sustainability, it is also about sort of driving better supply chain balancing and the optimisation of that. The whole idea of this really was to give the UK industry a leg up and prove that in real time. The other thing is that often when we’re getting a piece of data and we’ll go get it again, two years later. This solution is in real time so we can prove in real time that we know what’s in our supply chain, who our suppliers are. And, then we can then start layering in water data, labour data, wage data, diversity data, carbon data on top of those pieces, to actually have a real view of what it is that’s going on. That that then plays into things like the Higg Index, which you could layer on top as well. So what we can then start to show the world is actually we’ve got a truly repeatable, sustainable, transparent supply chain, and that we are taking a lead as far as that’s concerned.
What do you see as the next steps? How quickly could we see this roll out in a significant way?
We’re already defining exactly sort of what the next steps are with the partner organisations we worked with in the first place. This means how we can take the proof of concepts and rapidly take it to scale.
The conversations which are going on now, because we can already show we can do this technically, are about turning that into a commercial proposition and kind of getting it into the market relatively quickly. What we’re aiming at is that, within this quarter, that we will be able to kind of share exactly what the tool can do at scale, with a number of suppliers and in real time. We’re talking to a lot of different organisations around getting them engaged and we’re also looking to pull together a consortium of smaller organisations to then treat them as kind of a whole that we can then kind of engage with as well from that side of things. The scaling out process is purely down to how quickly from sort of a commercial and contracting side of things that we can move, because from a technological and process perspective, we’re pretty much already there. We’re looking to kind of really accelerate it this year, to have significant volumes in the in the platform this year, to and then to be moving beyond cotton very quickly and into other products. What I would hope to be able to say, if you asked me this question in a year’s time, is that we’ve already got a significant portion of the UK supply chain on the platform. We will move as quickly as our partners are happy to work.