The cold efficiency of automation, of minimal workers and maximum output, could be inescapable for firms wishing to keep up with competition. But what will this technology mean for supply chains and the people who maintain them?
Imagine a world where products are designed and manufactured for you at the click of a button. Say you discover a passion for running and want a pair of trainers that support your body and your progress. You use an app to select the style, colours, make sure the laces and soles match your workout gear, choose whether they’ll be vegan, and scan in a photo of your feet for the perfect fit. Once your order is sent, the system ensures the materials are in stock at a factory near your location, so the exact trainers you created can be shipped to your door in a matter of days. It may seem like a vision of the future, but sportswear giant Nike launched this direct-to-consumer model in 2020.
Sophisticated digital tools have been seeping into the everyday for the past few decades, often used to make a specific task quicker and easier. Procurement, like so many professions, is a mix not so much of the sacred and the profane as of the functional and the strategic. But there’s no denying that at its heart, it’s a series of complicated and deeply interconnected processes – from the interactions and decisions that drive purchasing through to the intricacies of the supplier relationship, risk mitigation and the monitoring and management of global supply chains. If we accept that any process can either be fully automated or at least significantly streamlined and improved by an element of automation, perhaps through AI, we should be asking an important question: can procurement itself be automated?
Going direct to consumers
Employees in many sectors have had to cope with the march of progress in the form of physical robots or robot process automation (RPA) encroaching on their jobs. Despite the investment required to set up such a system, automation is fast becoming the most effective way for companies to control costs and ensure quality. But as well as replacing hands, automated systems also feed information into the business brain, connecting with the Internet of Things, Big Data, AI and analysis systems to power deeper insights. According to a PwC survey in May 2020, 37% of CFOs are currently targeting autonomous technology roll-outs, and 34% aim to use this capability to improve decision-making and business agility – a mindset heavily influenced by the impacts of the pandemic. PwC US chairman and senior partner Tim Ryan said the survey showed a shift among executives to “focusing on what they can control”.
And control is certainly a driving factor at Nike. The company’s custom product service under its Consumer Direct Offense was three years in the making, and it’s paying off in savings and sales. With an initial goal of reducing order lead times from 60 days to 10, Nike reorganised its supply chain and operations through a combination of nearshoring and setting up small manufacturing facilities close to its key customer markets. It also developed and deployed 1,200 additive manufacturing machines, producing a network of highly automated facilities which reduced the number of steps in the manufacturing process by about 30%, and labour requirements by 50%. That’s quite a feat for a company with around 1m factory employees.
These figures were quoted as part of the company’s 1997 strategy, but when actually implemented 20 years later, building a highly automated manufacturing facility in Oregon in 2017, it found it could reduce staffing even further. “The original supply chain handling the supply for that market had 120,000 people,” explains Professor Richard Wilding of Cranfield University, “but this new facility, which did everything the old supply chain did, only needed 1,000 people.” Confidence was high and over the years Nike dropped almost all of its third-party distributors, including majors like Amazon.
This system enables direct engagement with consumers through various mobile apps including fitness programmes, health trackers and ecommerce platforms, and a key tool in its arsenal is Nike Fit, a 3D scanning technology bought from creator company Invertex. The software helps users to scan their feet so augmented reality images determine size and shape of the shoes, which can differ by style and sport type, instead of relying on off-the-peg averages. According to Nike, eradicating standard sizings is the company’s “ultimate goal”, which is not only a benefit to the customer but “a more accurate fit can contribute to everything from less shipping and fewer returns to better performance”. An added benefit is the perception of a bespoke service. While other businesses have looked to standardisation as a way to streamline supply chains and allow for greater automation, Nike has limited the number of products currently available in this measure-and-make way to keep it both personal and profitable.
As a digital disruptor, Nike certainly is in pole position having added automation to ordering, manufacturing and warehousing, as well as making numerous investments in technology startups. Last year, its DTC sales rose to 33.1% and in January 2021 the company reported sales 82% higher than the previous year – not bad results during a pandemic when many stores were shut or inaccessible. But let’s not forget this is the culmination of more than 30 years of work; moving forward, such transformations will surely be far quicker. “Digitisation and online retail are likely to be major factors in supply chain automation,” says Wilding, adding, “It’s something we’ve already started to see in the changes brought about by the coronavirus pandemic, as many companies shift to online operations out of necessity during lockdown. What may have taken years has been implemented in months as companies pivoted to survive the last year.”
Automation adds value
Control, efficiency through reduced labour and sustainability are cited as some of the top reasons leaders will prioritise automation in supply chains from this year, as is recouping lost revenues from the past 12 months. Prime areas are wherever repetitive or manual tasks are involved, and according to McKinsey, this correlates with higher degrees of physical proximity so it can also safeguard against transmission of illnesses like Covid-19. This goes a way to explaining why manufacturing, which is especially receptive to technological advancements, is the de facto sector people refer to when discussing automation. But this is closely followed by administration. For procurement teams, the admin involved in purchasing and vendor management tasks including data entry, form filling, calculations, information transfer and validation, could all be performed by autonomous systems.
Bosch, for example, is introducing autonomous systems to its 37,000 employee-strong supply chain management organisation. As well as making its own brand of household appliances, the electronics company also supplies electrical components to the automotive and aerospace sectors, among others. It has 26,500 suppliers and transports 300 million parts every day; the company’s 9,000-strong procurement team uses digital systems to make processes faster, more resource-efficient and transparent. It expects that by the end of this year, up to 85% of its global purchasing volume will be handled by cloud and platform solutions, such as the SupplyOn software platform designed for data acquisition, communication and exchange.
Munich-based SupplyOn is a joint venture between Bosch, Continental, Schaeffler and ZF, and enthusiasm for automated procurement is evidenced by the programme’s uptake by firms including Airbus, BMW, Liebherr, Safran, Schindler, Siemens and Thales. “Supply chain management at Bosch combines purchasing and logistics,” explains Arne Fleming, senior vice-president for supply chain management. “Close collaboration between these two functions and our partners from manufacturing allows us to shape the supply chain. Doing so reveals tremendous potential to leverage synergies and save costs.”
Siemens, another major cross-sector German conglomerate, is also embracing supply chain automation. Chief procurement officer Klaus Staubitzer is a proponent of the idea that it is a means to drive transformation. “I believe in technologies like artificial intelligence and robotics process automation,” he says. “There are technical capabilities available to us today that promise a great deal for our supply chain management because they create transparency and genuinely make a buyer’s life easier. What we have to do now is be clever in how we use these opportunities presented to us by the new technology.”
Siemens uses an in-house cost and value engineering system to establish overheads and costs drivers that feed into how its suppliers arrive at their prices. According to Staubitzer, it helps buyers add value to product development, in a way which is often “very surprising” to engineers. However, making suggestions that would reduce costs without affecting function or quality requires procurement to acquire additional knowledge of materials and terminology to hold such conversations. Otherwise, Staubitzer says, the engineers would insist on checking prices themselves. “If, however, the design engineer can see the buyer concerned can and should be of assistance with costing and even with the design process, the relationship that develops will be quite different and entail a much higher level of mutual acceptance.”
Robots bring fulfilment
Alongside these labour-intensive areas, warehousing and fulfilment are a natural fit for automation. However, retail rivals Walmart and Target have had somewhat tumultuous relationships with the tech, with both companies struggling to strike a balance between people and robotics over the past few years. For instance, in January 2020, Walmart extended its use of autonomous bots to 1,000 of its stores, having used them since 2017 to scan products on shelves for price accuracy and better manage stock and inventory levels in order to speed up fulfilment. But Walmart later scrapped the bots that November. And after trialling a shopfloor bot called the Tally, Target was unsure of the benefits and decided to stick with humans for now, but to automate the cleaning and cash counting duties.
This could be due to the bricks and mortar effect. Fleets of autonomous bots and inventory management systems could help smaller stores compete in terms of delivery times, but they are less economical in small shops, a place where customers often come for the human interaction. In contrast, online retailer Ocado has no physical stores and last year successfully trialled using robotics for pick and pack, inventory and fulfilment at its warehouses, following a dramatic upturn in orders due to the pandemic. Factories and warehouses have also been key areas for Nike, through deployment of autonomous goods-to-person robots, which enabled it to meet next-day delivery commitments. In stark contrast, sportswear rival Adidas took a mis-step with its entirely robotic manufacturing operation, Speedfactory. The factories, intended to negate the need for cheap labour, all closed down within three years.
The autonomous heavyweight
Of course, it’s impossible to talk about warehouse and fulfilment technology without mentioning Amazon. Many retailers benchmark their delivery targets by its standards, and for those not directly competing, its influence will still affect their operations as next-day is no longer good enough for consumers when Amazon can hit back with same-day and even within one hour. It seems virtually impossible for companies to compete without embracing automation.
As well as consumer influence, Amazon employs hundreds of thousands of people at its warehouses and distribution centres, all of whom its investments in technologies could make redundant. It reportedly owns over 200,000 robots that can scan barcodes as stock arrives, assigning it to the correct locations; transport goods to stowers who shelve them, scan them, and send them off to order pickers; and report on inventory levels and trends. In effect, the role of robot and human has switched, with workers now stationary and being constantly brought shelves of items by roaming robots. Progress never stops and Amazon is trialling autonomous packing machines, arms to load them, and shuttle bots to arrange stock in mini store rooms, reducing the opportunities for humans even further.
The company also plans to unveil its biggest ever fulfilment centre in Western Sydney, Australia at the end of this year, which at 200,000m2 will include its latest autonomous robotic systems. “The Amazon robotics fulfilment centre will more than double our operational footprint in Australia, enhance efficiency and safety for our associates while ultimately providing customers with wider selection and faster delivery,” Amazon Australia director of operations Craig Fuller has said.
It would appear that the applications for automation are myriad; governed by the tasks and not business type. However, it can be hard to assert its relevance to some – for instance, the construction sector, where logistics are complicated by the level of access to active sites and even simple administrative processes are harder to change. “Automated purchasing isn’t a standard term in the construction sector,” said Will Hughes, emeritus professor of construction management and economics at the University of Reading.
“It’s a standard term in organisational purchasing, of things businesses need, but construction is unlike anything else an organisation might buy. It involves a huge amount of money and one duff project could bankrupt the contractor or the client. There are fundamental characteristics to do with scale, scope and complexity that make construction quite unlike other industrial sectors. It’s not that it’s lagging behind other sectors in terms of technologies, it’s that it’s a very special case that’s difficult to deal with.”
An important difference Hughes highlights is that even just-in-time delivery, standard practice in much of the profession, is almost impossible to apply in construction projects. “You can’t control the road network and other parts of the transportation network so you need storage onsite, and that will have to handle buffer stocks to ensure the project’s labour resources can be permanently busy. And those labour resources are highly specialised and they usually can’t be switched around because construction has to proceed in a fixed order. You can’t switch skilled door hangers to installing floorboards or window frames, and plasterers can’t do electrical work.”
One area in which construction firms are managing to make transformations is in modular construction, where large sections of buildings are produced and finished at off-site factories then transported to the site for a simpler assembly process. These factories can be highly automated and don’t have the access problems of building sites, so more advanced supply chain processes can be used. Unsurprisingly, Hughes says, Scandinavia is leading the way in this shift.
“It’s something you see particularly in the northern regions, outside the cities, where there is extensive use of cross-laminated wood in buildings. The ground work is always site-specific, but once you’ve got a flat concrete slab everything else can be effectively slotted together like a flat pack. IKEA, in fact, is heavily involved in this.” This is transformative for the construction industry, because the in-factory manufacturing of modules allows a great deal of customisation. But for now, construction is one of the few industries relatively untouched by automation.
Will robots take our jobs?
So let’s address the elephant in the room. Will automation decimate the future of procurement and supply chain jobs? According to a McKinsey estimate, the appetite for post-pandemic automation will result in 25% more global employees being forced to change their jobs. But while figures such as these lead to fear of obsolescence and a dystopian future where humans have no work, the reality is that technology will necessitate a shift to different skillsets.
In India, the government drive to accelerate automation adoption is actually rooted in the potential for job creation and enriching society as a whole. In 2020, an Automation Everywhere and EY joint study said India’s GDP could gain $365bn from wider use of automation, bringing total GDP to $5tn by 2024-2025. “As more organisations adopt intelligent automation at scale, this will eventually create more high-skilled jobs, providing more economic opportunities for everyone,” Kamalanand Nithianandan, Partner, Advisory Services, EY India said.
In contrast, research consultancy Forrester’s vision is relatively bleak, forecasting that Australia’s job market will shrink by 11% by 2030 due to automation. The company warns that procurement, along with finance and accounting, will be the first hit, losing one million roles in the coming years; however, it says this same technology will also create 1.7 million new jobs, so it’s not all doom and gloom. While repetitive, labour-intensive tasks will be automated, remaining procurement roles are likely to fall into two distinct factions: digital prowess and emotional intelligence. To avoid becoming what principal analyst Sam Higgins calls a digital outcast, Forrester says a focus on “superior human physical communication abilities and empathy” would safeguard 27% of jobs in Australia, and help grow a bank of “digital elites”.
For a preview of how this might play out in reality, we can examine another profession that is process-based but also highly strategic and dependent on human judgment. The legal profession feels less susceptible to algorithmic innovation than procurement, since much of its work (think lawyers pleading for their clients in court) is so subjective. In truth, however, it is highly based on the use of precedent, the application of tightly defined parameters and the crunching of information.
That’s prime territory for AI. JP Morgan unveiled software in 2017 that it claimed could scan documentation so rapidly it could perform analytical tasks in seconds that would take the average paralegal 360,000 hours. Most sizeable law firms have begun to introduce AI capabilities into their operations, albeit behind the scenes, and the effects are already being felt: while it is now commonplace for leading London lawyers to charge £1,000 per hour, the Institute for Employment Studies thinks 35,000 UK jobs in the sector will be lost in the next 10 years. The impact will be felt among clerks, juniors and auxiliary staff whose jobs can be performed more efficiently by software.
Where human judgment is required, human lawyers (particularly at the top end) will still be in work. In any automated process, lack of emotional intelligence is a stumbling block: as Scott Dance, director of Hays Procurement and Supply Chain says, no automated system could have predicted that this time last year all the supermarkets would run out of toilet rolls. Dance says supply chain management courses will soon start to target those vulnerable areas of technology – negotiating and relationship management – as well as data manipulation.
“For an entry-level role at the moment, you get a procurement analyst to come and analyse your supply chain, how much you spend with them, who you use, doing a demand plan and forecasting – you could get a computer to do that. But then what you need to manage is a) what is put into that system (because data is only as good as what you put into it) which normally requires human interaction at the beginning. Then at the end, once you’ve got that data, ask how can I use that to gain competitive advantage for my business? Again, that is where a human will come in.”
Another associated change is that the executive level in businesses will have to take a greater role in supply chain decisions. PwC’s vice-chair and chief clients officer Amity Millhiser expects to see a focus on automating data collection and analysis regarding supply chain effectiveness, and leaders will gain stronger presence and influence over their organisations, based on the in-depth, real-world, real-time data autonomous systems will provide. They will likely lead teams that are smaller and more specialised. “If customer interaction moves online, we have problems with what the high street looks like, and associated problems with social isolation. Most procurement and supply chain decisions are not taken at board-level, they are for middle management. All these things need to start becoming board-level discussions,” says Wilding.
“For example,” he adds, “if a supermarket said they would not fly in roses for Valentine’s Day from Africa, they could force towns which depend on that trade into destitution. So then you start having to look at different ways, better ways, to support those communities. You might be using data from satellites to evaluate the environmental performance of suppliers, but decisions based on those data will have repercussions that need to be examined at a high level, and by humans.”
A blended workforce
However, as Nike and Amazon demonstrate, the path to automation in certain sectors is inextricably linked to smaller workforces and alternative roles for people, which is promising if your future lies in robot maintenance or software development, but less so in purchasing. For others, blended teams of human+system are more likely, therefore managing transition periods and properly engaging staff is of the essence.
As Staubitzer warns, companies pursuing supply chain automation should be careful and sensitive of how they implement it because “digitisation can become a burden if it does not ultimately make life for your people and your organisation clearer, simpler and better. And if it becomes a burden, it will not be accepted and can never really work”. Siemens mitigated this in its own tranformation project by first asking employees to consider how digitisation could be put to good use in their setting and what would make sense for the business. The management team monitors projects and helps teams keep a handle on any issues that arise, which prevents workers feeling that the systems have been imposed upon them from above.
This approach is also recommended by PwC’s Suneet Dua, principal, chief product officer, who emphasises the importance of integrating the technology so it acts as an enabler and not a hindrance or replacement. “People have to make adoption of automation the fabric of their job,” says Dua. “Adoption is the new engagement margin. And what we mean by that is if people don’t adopt the automation technologies you provide, then you’re not going to change the business.”
Across customer service, purchasing, contract administration, fulfilment and distribution, automation promises radical changes for the profession, and businesses need to keep up with both competitors and consumer demands, or else they risk falling behind fast. According to Gartner, almost a quarter (23%) of supply chain leaders expect to have a digital ecosystem in place by 2025, compared with 1% today – that’s a huge jump in just four years. Gartner says this is largely motivated by the level of disruption caused by the pandemic, as many leaders believe digitisation, including adoption of the Internet of Things, blockchain, digital twins and automation, will improve business resilience and protect against disruptions in the decades ahead.
All companies are aiming to improve their resilience and while efforts in digitisation have been inconsistent and it is often treated as a nice-to-have, trends suggest we’re heading for a sudden change of pace, as companies try to keep up with the competition. The rewards are big and the potential losses of falling behind are even bigger. But regardless of how expansive autonomous systems become, procurement has interpersonal touchpoints that technology simply cannot replace – at least for the foreseeable future – making humans not redundant but a precious resource. So invest in soft skills, and if you’re wondering whether this could really happen, if procurement could some day become automated, simply look around you. It already is.