Supply Chain 4.0 – the application of the Internet of Things, the use of advanced robotics, and the application of advanced analytics of big data in supply chain management: place sensors in everything create networks everywhere, automate anything, and analyze everything to significantly improve performance and customer satisfaction. Over the last thirty years, logistics has undergone a tremendous change: from a purely operational function that reported to sales or manufacturing and focused on ensuring the supply of production lines and the delivery to customers, to an independent supply chain management function that in some companies is already being led by a CSO – the Chief Supply Chain Officer. The focus of the supply chain management function has shifted to advanced planning processes, such as analytical demand planning or integrated S&OP, which have become established business processes in many companies, while operational logistics has often been outsourced to third-party LSPs. The supply chain function ensures integrated operations from customers to suppliers.
Trends in supply chain management
Industry 4.0 creates a disruption and requires companies to rethink the way they design their supply chain. Several technologies have emerged that are altering traditional ways of working. On top of this, megatrends and customer expectations change the game. Besides the need to adapt, supply chains also have the opportunity to reach the next horizon of operational effectiveness, leverage emerging digital supply chain business models, and transform the company into a digital supply chain. Several megatrends have a heavy influence on supply chain management: there is a continuing growth of the rural areas worldwide, with wealth shifting into regions that have not been served before. Pressure to reduce carbon emissions as well as regulations of traffic for socioeconomic reasons adds to the challenges that logistics are facing. But changing demographics also lead to reduced labor availability as well as increasing ergonomic requirements that arise as the workforce age increases.
At the same time, customer expectations are growing: the online trend of the last few years has led to increasing service expectations combined with a much stronger granularization of orders. There is also a very definite trend towards further individualization and customization that drives the strong growth of and constant changes in the SKU portfolio. The online-enabled transparency and easy access to a multitude of options regarding where to shop and what to buy drive the competition in supply chains.
The vision of the future state
The digitization of the supply chain enables companies to address the new requirements of the customers, the challenges on the supply side as well as the remaining expectations for efficiency improvement. Digitization brings about a Supply Chain 4.0, which will be …
New product distribution approaches reduce high runners’ delivery time to a few hours. The basis for these services is built by advanced forecasting approaches, e.g., predictive analytics of internal (e.g., demand) and external (e.g., market trends, weather, school vacation, construction indices) data as well as machine status data for spare-parts demand, and provides a much more precise forecast of customer demand. Forecasts are not carried out on a monthly basis, but weekly, and for the very fast-moving products even every day. In the future, we will see “predictive shipping,” for which Amazon holds a patent – products are shipped before the customer places an order. The customer order is later on matched with a shipment that is already in the logistics network (being transported towards the customer region) and the shipment is rerouted to the exact customer destination.
… more flexible.
Ad hoc and real-time planning allow a flexible reaction to changing demand or supply situations. Planning cycles and frozen periods are minimized and planning becomes a continuous process that is able to react dynamically to changing requirements or constraints (e.g., real-time production capacity feedback from machines). Once the products are sent, increased flexibility in the delivery processes allows customers to reroute shipments to the most convenient destination.
New business models, such as Supply Chain as a Service for supply chain planning functions or transport management, increase flexibility in the supply chain organization. The supply chain can be bought as a service and paid for on a by-usage basis instead of having the resources and capabilities in-house. The specialization and focus of service providers allow them to create economies of scale as well as economies of scope and also attractive outsourcing opportunities.
For example, we will see an “Uberization” of transport: crowd-sourced, flexible transport capacity, which will lead to a significant increase in agility in distribution networks.
… more granular.
The demand customers for more and more individualized products is continuously increasing. That gives a strong push towards micro-segmentation, and mass customization ideas will finally be implemented. Customers are managed in much more granular clusters and a broad spectrum of suited products will be offered. This enables customers to select one of multiple “logistics menus” that exactly fits their need.
New transport concepts, such as drone delivery, allow companies to manage the last mile efficiently for single and high-value dense packages.
… more accurate.
The next generation of performance management systems provides real-time, end-to-end transparency throughout the supply chain. The span of information reaches from synthesized top-level KPIs, such as overall service level, to very granular process data, such as the exact position of trucks in the network. This range of data provides a joint information basis for all levels of seniority and functions in the supply chain. The integration of data of suppliers, service providers, etc. in a “supply chain cloud” ensures that all stakeholders steer and decide based on the same facts.
In digital performance management systems, clean-sheet models for warehousing, transport, or inventory are used to set targets automatically. To keep the aspiration of targets also in case of supply chain disruptions, systems will automatically adjust targets that cannot be achieved anymore to a realistic aspiration level. We will see performance management systems that “learn” to automatically identify risks or exceptions and will change supply chain parameters in a closed-loop learning approach to mitigate them. That enables the automatic performance management control tower to handle a broad spectrum of exceptions without human involvement and to only leverage the human planner for disruptive events/new events – with this, a supply chain is continuously developing towards its efficient frontier.
… more efficient.
Efficiency in the supply chain is boosted by the automation of both physical tasks and planning. Robots handle the material (pallets/boxes as well as single pieces) completely automatically along the warehouse process – from receiving/unloading to putting away to picking, packing, and shipping. Autonomous trucks transport the products within the network. To optimize truck utilization and increase transport flexibility, cross-company transport optimization is applied to share capacities between companies. The network setup itself is continuously optimized to ensure an optimal fit to business requirements.
To create an ideal workload in the supply chain, various transparency and dynamic planning approaches are leveraged to drive advanced demand-shaping activities (e.g., special offers for delivery time slots with low truck utilization).
Digital waste prevents supply chains from leveraging the potential of Supply Chain 4.0
In today’s supply chains, many sources of digital waste can be found (in addition to the existing waste) that prevent the potential of Supply Chain 4.0. It is crucial to understand the sources of waste and develop solutions to reduce/avoid it in the future state. The sources of digital waste can be classified into three types:
1. Data capturing and management
Often, available data is handled manually (data collection in a system, paper-based data handling, etc.) and not updated regularly, e.g., master data on supplier lead time that is entered once (sometimes even only dummy numbers) and then remains unchanged for years. Another example in warehousing is advanced shipping notifications, which are received but not used to optimize the inbound process.
On top of these examples, it is typically not clear which additional data could be leveraged to improve processes, e.g., sensing of supply disruptions – if the lead time of a supplier is continuously increasing, a warning should be sent out to make planners aware of the situation and enable them to mitigate supply disruptions at an early stage. In current systems, this signal will not be recognized and will lead to a lower supplier service level reported at the end of the month. If the worst comes to the worst, the issue will cause trouble in the assembly line replenishment and operational problems.
2. Integrated process optimization
Many companies have started to implement an integrated planning process, but very often this is still done in silos and not all information is leveraged to achieve the best planning result possible. In addition, it can frequently be observed that automatically determined planning or statistical forecast data is manually overwritten by planners. Especially for parts moving at medium or high speed, manual overwrites usually have a negative impact on forecasting accuracy. Besides the intracompany optimization, the process optimization between companies has not been fully leveraged yet and improvement potentials created by increased transparency are not realized.
To get to the advanced level of integrated process optimization, the organizational setup, governance, processes, and incentives need to be aligned within and between partners in the supply chain.
3. Physical process execution of humans and machines
Nowadays, warehousing, assembly line replenishment, transport management, etc. is often done based on gut feeling, but not leveraging available data, e.g., to improve pick paths in the warehouse. Warehouse operations are still managed in batches of one to two hours, not allowing the real-time allocation of new orders and dynamic routing. Also, opportunities arising from new devices, such as wearables (e.g., Google Glass) or exoskeletons, are not leveraged.
Future supply chain planning will largely benefit from big data and advanced analytics as well as from the automation of knowledge work. Two example levers with significant impact are “predictive analytics in demand planning” and “closed-loop planning.”
Predictive analytics in demand planning analyzes hundreds to thousands of internal as well as external demand influencing variables (e.g., weather, trends from social networks, sensor data) with Bayesian network and machine learning approaches to uncover and model the complex relationships and derive an accurate and granular demand plan. These new technologies enable a significant improvement in demand forecast accuracy, often reducing the forecasting error by 30 to 50 percent. Also, the days of a “single truth” regarding the forecasting numbers are over – these advanced algorithms provide probability distributions of the expected demand volume rather than a single forecast number. This allows for targeted discussions, including upside potential and downside risks in the S&OPs, and advanced inventory management approaches.
Widely automated and fully integrated closed-loop demand and supply planning breaks the traditional boundaries between the different planning steps and transforms planning into a flexible, continuous process. Instead of using fixed safety stocks, each replenishment planning considers the expected demand probability distribution and replenishes to fulfill a certain service level – the resulting implicit safety stocks are therefore different with every single reorder. Another powerful feature of closed-loop planning is the integration of pricing decisions with demand and supply planning; depending on the stock levels, expected demand, and capability to replenish, prices can be dynamically adapted to optimize the overall profit made and minimize inventories at the same time.
Logistics will take a huge step change through better connectivity, advanced analytics, additive manufacturing, and advanced automation. For example, as warehouses are being automated, we will see a significantly increased amount of autonomous and smart vehicles, and 3-D printing changes warehousing and inventory management strategies completely.
The next generation of touch, voice, and graphical user interfaces and their quick proliferation via consumer devices facilitates much better integration of machines in almost any process in warehousing operations. For example, the breakthrough of optical head-mounted displays, such as Google Glass, enables location-based instructions to workers, giving guidance for the picking process. Advanced robotics solutions have emerged for the improved picking of cases and single pieces, and the use of exoskeletons (that emulate human physiology and can support training manual movements) will have a major impact on warehouse productivity. In total, warehouse automation become much more holistic, with some warehouses being fully linked to production loading points, so that the entire process is carried out without manual intervention.
Autonomous and smart vehicles will lead to significant operating cost reduction in transportation and product handling and at the same time provide benefits regarding lead times and lower environmental costs. The use of self-guided vehicles in controlled environments (e.g., mines) or on-premise solutions (e.g., trains) as well as AGVs in warehouse environments are already operational and will further grow significantly in the near future. Autonomous trucks for use on public streets, however, are just being piloted in Europe and North America with promising results so far.
Besides the automation of warehouse processes, additive manufacturing will also have a significant impact on physical flows in the supply chain. For example, 3-D printing has become much more relevant for a broad range of business applications, such as the local production of slowly moving spare parts or tools. This development is driven by an expanding range of printing materials, rapidly declining prices for printers, and increased precision and quality. By now, the first production facilities that operate exclusively with 3-D printers have been established.
Performance management is indeed changing tremendously. Whereas in the past, the generation of KPI dashboards was a major task and KPIs were only available at aggregated levels, now granular data is available in real-time from internal and external sources. This moves the performance management process from a regular, often monthly process to an operational process aimed at exception handling and continuous improvement. For example, planners can be pointed to critical supply chain disruptions and further supported by the automatic handling of minor exceptions or potential solutions for the larger ones.
Automated root cause analyses are one approach for exception handling. The performance management system is able to identify the root causes of an exception by either comparing it to a predefined set of underlying indicators or by conducting big data analyses, leveraging data mining and machine learning techniques. Based on the identified root cause, the system will automatically trigger countermeasures, such as activating a replenishment order or changing parameter settings in the planning systems, such as safety stocks.
Two examples of how order management is improved are no-touch order processing and real-time replanning, which lead to lower costs through automation of efforts, higher reliability due to granular feedback, and superior customer experience through immediate and reliable responses.
No-touch order processing is the logical next step after implementing a reliable available-to-promise (ATP) process. Through an integration of the ordering systems, linking to ATP, and enrichment with order rules, the system can be used to fully automate the ordering process. The goal is to have a complete “no-touch” process, where no manual intervention is required between order intake and order confirmation. Very stringent order rules that have to be followed, and continuously updated master data are prerequisites.
Real-time replanning enables order date confirmations through instantaneous, in-memory replanning of the production schedule and replenishment in consideration of all constraints. Therefore the supply chain setup is always up to date, leading to a very reliable planning base. On top, additional services can be offered to the customers, e.g., a faster lead time for a certain premium fee, so the customer can see the feasibility and the updated dates at a glance.
The supply chain cloud forms the next level of collaboration in the supply chain. Supply chain clouds are joint supply chain platforms between customers, the company, and suppliers, providing either a shared logistics infrastructure or even joint planning solutions. Especially in noncompetitive relationships, partners can decide to tackle supply chain tasks together to save admin costs, and also to leverage best practices, and learn from each other.
Another major field within a collaboration is end-to-end/multitier connectivity. Where some automotive companies have already started collaborating throughout the entire value chain (e.g., from the cow farmer to the finished leather seat in the car), other companies still need to close this gap. The collaboration along the value chain allows for overall much lower inventories through an exchange of reliable planning data, a step change in lead time reduction through instantaneous information provision throughout the entire chain, and an early-warning system and the ability to react fast to disruptions anywhere.
Supply chain strategy
Following the need for further individualization and customization of the supply chain, supply chain setups adopt many more segments. To excel in this setting, supply chains must master “micro-segmentation.” The granularization of the supply chain into hundreds of individual supply chain segments based on customer requirements and own capabilities designed in a dynamic, big data approach allows for mass-customized supply chain offerings. Tailored products provide optimal value for the customer and help minimize costs and inventory in the supply chain.
Author: Danish Mairaj, CISCOM, PMP