Leaders of businesses large and small are asking their teams to start automating their processes and workflows. The question is, where to start? 

A collection of tools and processes that are the basis for automation must first be created. Developing this framework is key to moving away from siloed functions and toward seamless, synchronized operations that orchestrate automation across the supply chain. An autonomous supply chain adapts and reacts to disruption by providing visibility across systems, streamlining workflows, eliminating silos and enabling autonomous disruption management.  

Before that framework can be built, all the input — including assets, transactions, processes, etc. — must be digitalized.  

Many larger organizations have undertaken or largely completed that task. Medium and small businesses are likely not as far along. Even if much of their information is in the form of spreadsheets and/or emails, they are left with siloed, manual and often redundant processes. Today, systems are emerging that can digitalize those information streams by leveraging algorithms and expanding them with artificial intelligence (AI) and machine learning (ML). 

If it seems this might be daunting for smaller companies, look at start-ups that have experienced tremendous success, such as today’s global rideshare services, whose digital platforms were essential to their growth.  

This report explores the impacts autonomy across the supply chain is having on manufacturers, retailers and logistics service providers (LSPs), and outlines steps for building an automation framework. 

Improving Supply Chain Basics 

Automation, going well beyond robotics, impacts a wide range of supply chain processes and outcomes, including adaptability and resiliency, operational efficiency and customer experience. 

Roy Bridgland, senior industry strategies director, Blue Yonder, sees it this way: “We need to take a broader view of automation, by looking at the processes themselves. Automation can deliver streamlined intelligent autonomous decision-making processes way before the robot can. It's that sort of automation that is affecting the actions and tasks that the robot is performing in the first place.” 

Adaptability and resiliency are critical to successful supply chain design, but have been difficult to achieve in the complex and dynamic realities of global operations. Automation inherently improves real-world results with nearly instantaneous capabilities to identify, respond to and recover from disruptions, yielding much more adaptable, resilient services. 

Adaptability and resiliency also support greater operational efficiency and productivity, as automation continuously improves asset and resource utilization, which inevitably leads to cost reduction. 

Customers take favorable notice of faster delivery times that result from more efficient and accurate fulfillment processes. Simultaneously, with digitalized shipment information, customers are experiencing much better visibility, and are gaining confidence that resources and goods will arrive when expected. 

First Steps 

For small- to medium-size shippers or LSPs who accept orders by email, phone and other traditional channels, automation can make quick impacts in basic operations. These include grouping shipments into a load for a truck, or transitioning from a paper picking list, and transmitting those tasks to individuals who are using handheld terminals that provide real-time updates as they work. 

This sets the stage for a shipper to optimize operations, using more system rules. In the instance of a truckload, rather than simply grouping shipments because they're going to the same geographic area, the automated system considers the journey and the distance to then determine if a destination-based load is really the optimal grouping of shipments. 

As AI and ML capabilities gain experience, they can drive more accurate and faster predictions.  

According to Greg Sloyer, manufacturing and industry principal with AI data cloud platform Snowflake, “Forecasts can be wrong to some extent. Now, however, we can react in real time by revealing the causes of forecast errors. With the analytics available today we have better data-driven insights that can instantly identify supply interruptions, changing market demands and other critical variables, and automatically react.” 

With time, supply chain operations will be more “hands-off.”  The decision-making software will select carriers, create loads and transmit details to customers. Human guidance will be requested when conditions arise that the system isn’t prepared to handle. In such cases, options will usually be presented for a fast and well-informed decision. 

Automation on the Edge 

Automated systems usually live in the cloud, but when connected to “edge” technologies such as automated mechanical or robotic systems, or data-capture devices such as sensors, cameras or scanners, they form an automated sub-system which operates locally.  

As an example, cameras have been in wide industry use for many years, but mostly for security. Now, automated camera systems can read images at a location to monitor traffic flows, identify equipment in the yard, and inform other process-management needs.  

The data from these devices can be processed on site by an automated camera sub-system, which interacts seamlessly with the wider, cloud-based automated ecosystem.  

In the first phase of deploying automated camera technology, the systems use ML to recognize trailer IDs for identification in the yard. This learning is not instantaneous, but a process during which the accuracy of the recognition improves. Once reliable, that data can be linked to the warehouse system to determine the priority of the delivery and to direct trucks to the yard, a dock door, or some other location. 

Once the system is trained at an initial location, this process can be quickly distributed to, and implemented at, multiple locations for rapid enterprise-wide adoption. 

Intelligent and Independent Decision-Making 

Full implementation of autonomous decision-making is already underway at some companies. 

Such is the case at one of the world’s leading shipping and logistics providers, which has been an early adopter of advanced supply chain technology in hundreds of warehouses worldwide. 

It has been using radio frequency handheld devices for picking orders for many years. That digitalized information is now optimized by the warehouse management system (WMS), which groups the right orders together for the most efficient route around the warehouse. 

The next phase in their automation process was the introduction of mobile robots linked to a warehouse execution system that coordinates the WMS with the robotic system. As robot vendors were added to this system, the speed of implementing the system at other warehouses increased rapidly because each vendor’s robots were pre-programmed.  

Now the company is working to enable robots from multiple vendors to communicate with one another — backwards and forwards — between the bots and the WMS. Working with a provider of autonomous robotics solutions, rapid progress is being made toward a cloud-based solution that provides immediate access to multiple robot vendors whose machines can “plug and play” off the shelf. 

The goal is to make robots rapidly adaptable to different situations. These would include entering a warehouse to find and bring out stock, traveling and stopping at multiple locations where associates load stock into them, or other scenarios. Each one of these will be held in a pre-loaded library in the system, and made available to all new and existing system users.                                       

Optimization, Digitalization, and Cognitive Execution 

Beyond robots, automation can affect profound changes in virtually every aspect of a supply chain, including digitalization, optimization and cognitive execution. 

Digitalization is the basic process of converting data and workflows into a digital format. It’s the first step on any journey to building an automation framework. Without taking this step, information can’t be processed in an autonomous way.  

An example is the paper pick list. Once it’s in digital form, those tasks are downloaded to an edge device that workers can use to confirm they've picked an item. The digitalized workflow is now started, and elements such as pick order can be optimized for more efficient work, picking tasks prioritized based on order or transportation needs, and labor and robotics interleaved in order to complete tasks more effectively. 

Optimization has been evolving over time. A current static model can analyze multiple options, but only within the constraints of limited parameters and the statistical model that it's been programmed to work around. Within that scope it drives toward the least cost, performs transportation route planning, etc.  

Historically, it couldn’t dynamically act on information received in real time, such as when a vehicle encounters a traffic jam or other disruptions. 

“Now, with AI and ML,” Bridgland says, “a supercharged optimization model can take in the information around it, it's able to decide on its own, and it's able to create another task or create a secondary action. Using real-world events, it can update its decisions to create a new workflow or adjust priorities of tasks that might involve some tradeoffs but are still optimized decisions.” 

With digitalization and automated optimization, the next step is to initiate cognitive execution. By applying AI and ML to analyze relevant data about factors — both current and historical — a much broader data set becomes useful. Cognitive execution can recalculate, and then take a corrected path towards achieving the original objective. 

If there is a warehouse labor shortage for example, autonomous systems can dynamically switch orders to another warehouse, or update the orders and routes to maximize what is going through the robot fleet versus the manual process.  

Finding Solutions  

With an understanding of the tremendous potential for building an automation framework for more competitive supply chain operations, it’s time to get started. 

Manufacturers, retailers, and LSPs are entering a new era of competition to enhance profitability and drive customer focus. Costs to serve, absenteeism and customer demands remain high, and continue to grow. The situation is compounded by higher inflation, interest rates, fuel costs and labor costs.  

Organizations can now streamline operations, build adaptive strategies, improve margins and increase customer centricity. By emphasizing digitalization, predictiveness and autonomy, new levels of success in speed, accuracy, service and revenue can be realized. 

An accelerated and autonomous supply chain adapts and reacts by providing visibility across systems, streamlining workflows, eliminating silos, and enabling autonomous disruption management.  

By identifying the synergies across execution systems, a greater outcome is achieved than is possible with the sum of the individual parts, resulting in a highly customer-centric, efficient, resilient, agile and connected supply chain. 

It’s key to find a partner with capabilities that will impact any or all of your critical systems including: 

              •Order management

              •Warehouse management

              •Warehouse execution

              •Yard management

              •Transportation management

              •Transportation planning and optimization 

Start by taking the basic steps to produce positive, measurable initial results that will lead to a commitment across the enterprise for a continued journey down a logical, manageable path toward full supply chain automation. 

Resource Link: now.blueyonder.com/automation-acceleration-mfg.html