Supply chain control towers have evolved from mere visibility tools to intelligent planning and autonomous execution systems with prescriptive analytics and autonomous agents. So, what exactly is a control tower? Put simply, a control tower is a supply chain management system that embodies real-time visibility, decision making, and execution capabilities across the end-to-end supply chain. It enables views and actions on orders and shipments down to individual items, and allows collaboration across all participants. To be effective, a control tower should run on real-time data in a single shared database and be available to any and all trading partners in the supply chain. Today’s advanced control towers go far beyond visibility, providing predictive and prescriptive analytics for complex problem solving and optimization at the network level.
For companies worldwide, recent events have exposed enormous gaps in supply chain systems and processes, as companies struggled to respond to supply and demand disruptions. They’ve often faced the inability to answer basic questions, such as: Where is my shipment and when will it arrive? What’s the hold-up and which customer orders are impacted? To more difficult questions such as: How do I optimize my decisions across the region, when all my decisions are so interconnected? How can I offload routine chores, so I have more time to focus on what matters?
Why Networks Are The Foundation for Smart Control Towers
There are many ways to implement control towers, and results will vary based on how it’s done. Modern, network-based control towers operate on a single database – a single version of the truth – with all services orchestrated intelligently across the suppliers, customers, carriers, co-manufacturers, and distributors in your business network. This is a key differentiator in a network-based strategy.
This is important, because when all parties and systems are connected to the network, it enables optimization and decision-making based on real-time information and takes into account the capacities and constraints across the network. This means better planning and better execution and ensures the highest service levels at the lowest landed cost. It’s only this network-level view that is able to consider the complex cost and service level tradeoffs for situations with hundreds or thousands of orders for businesses with potentially tens of thousands of SKUs.
This is why an effective control tower needs to be a real-time network application spanning the end-to-end supply chain, that includes all trading partners from the source of supply to the point of consumption.
As analyst firm ChainLink Research pointed out, “The new Networked Enterprise Model of the 21st century requires implementing a supply chain trading partner operating model that supports the networked enterprise, providing interoperability and visibility with reduced complexity. The supply chain trading partner operating model is a codification and automation of the key processes, policies, performance metrics and technology that govern your supply chain partner relationships.”
The Road to Smarter Control Towers – Benefits of a Network Approach
Providing a single version of the truth, this solid foundation of network-based connectivity enables:
- End-to-End and Real-Time Visibility. Bringing together parties, facilities, and transportation into a single network view, organizations have visibility to all events and milestones that they want to track.
To successfully represent the complete supply and logistics lifecycle across a network of supply chain participants (warehouses, plants, DCs, carriers, LSPs, customers, contract manufacturers, and suppliers), the supply chain control tower solution serves as a system of engagement, integrating and harmonizing data across internal and external players. It also operates with the capability to embrace many systems and to assume control of any point in the execution lifecycle as needed, possibly even serving to replace legacy or redundant systems. By integrating transaction information across all parties, the control tower solution enables real-time visibility of the entire supply and logistics lifecycle from purchase order to shipment order, shipment execution, track and trace, and to financial settlement.
- Performance Measurement, Predictive Analytics and Alerts. This initial end-to-end visibility is supported with secure permissions and dashboarding configurations that allow all participants to view and analyze in real-time the data relevant to their organization and user role.
Operational dashboards can be configured at the user level and include all the tasks and alerts relevant to each specific role. Using these operational dashboards, users can drill into each task and alert to get more information and take steps to resolve any issues.
Command and control monitoring rolls up data to provide a management overview of the health of the network. It should be configurable to allow for managers and executives to customize the KPIs to give them a unique perspective into the factors and performance that they care about, e.g., costs, service, financials, risk.
These kinds of network-supported reporting mechanisms provide unmatched opportunities to improve cost and service within the organization and across the network. But things start to get interesting when intelligent agents that run on top of the execution data are introduced. These algorithms can combine other streaming data sources and compare the plan in real-time and predict ETAs and other events. Then exceptions can be highlighted and managed and subjected to root-cause analytics, to help provide insight, predict service levels, and drive performance improvements.
- Decision Support, Recommendations, Case Management and Prescriptive Workflows. The next step is to leverage this data and these analytics to support human decision-making and workflows. With real-time end-to-end visibility and actionability across demand channels, supply sources, and all modes of transportation, the system can now continuously compare planned activities to actuals. It also determines the likely impact, alerts relevant parties of deviations, and recommends optimized solutions so that exceptions can be managed quickly and optimally.
A network-enabled control tower goes beyond what traditional supply chain and TMS solutions can accomplish, by allowing network participants to collaborate on time-sensitive issues in real-time. All partners can manage issues that occur during the execution process.
Prescriptive workflows can go even further to incorporate facilities, such as plants, warehouses, yard and dock scheduling operations, into the end-to-end workflow. Processes that are frequently managed across separate systems or via manual communication between Shipper, 3PL and Carrier, can be streamlined across all parties where specific roles, activities and performance metrics can be established for optimized operations.
- Autonomous Network Control. At this level, intelligent autonomous agents propose solutions, identify the best option, and execute it. This is an evolution. Many processes to suggest resolutions and user-managed actions can be gradually converted to automated network actions. Now the intelligent control tower goes beyond detecting trends to automating issue resolution. It can propose optimal solutions at the network level to improve operational efficiencies and maximize service levels.
A control tower on a real-time demand and supply network across all trading partners, surfaces many more problems than organizations have ever been able to handle in the past. By defining limits and boundaries for automated decision-making via “guardrails” organizations can automate decision-making within those “guardrails” to resolve problems as a normal course of business, but at much larger scale than ever before possible.
The autonomous agents include advanced planning and optimization algorithms that can autonomously make and execute decisions within the guardrails as well. Agents can do things like adjust and create orders, modify, or change inventory policies, forecast transport capacity, perform transport optimization for dedicated and common carrier equipment, as well as the advanced scheduling of DC appointments.
It’s important to understand, that this is an evolution of capabilities, it’s not an all-or-nothing proposition. It involves experimentation, evolution, gaining trust in the automation results, and human supervision, until agents are surpassing expectations. Then they can be turned on to run autonomously. Users can gradually cede autonomy to intelligent agents where it makes sense, and then only within certain products or parameters. As the system learns, improves, and proves itself, it gains more and more autonomy. Ultimately, the intelligent agents embedded within the execution layer can run the supply network with little human intervention.
Find Problems Faster, Solve Them Optimally
Control towers on this multiparty network architecture, not only expose many more problems in the network sooner – while there’s still time to fix them at lower cost – they also provide many more ways to solve them. When the architecture treats all variables in the system as dependent and being driven by the desired outcome, then the control tower offers a wide range of solutions for each and every problem in the network.
A multi-echelon supply network problem can be solved in many different ways, for example, by adding shifts, expediting, allocations, postponement, etc. In fact, certain problems can have 4 or 5 different solutions involving many different parts of the supply network, each having a different business effect on the network outcome. The AI engine can rank these choices based on setting various business strategies for market share vs revenue vs margin generation vs customer service vs resiliency, etc.
What about those that fall outside the guardrail “safety zone”? These more difficult problems can be elevated to expert users via optimization workbenches powered by predictive and prescriptive analytics in the control tower. Here human planners, schedulers, distributors, shippers, operators, manufacturers, etc., can review all the factors and agent-generated recommendations before executing a solution and/or collaborating with partners to explore other options. Workbenches need to support various functions across the supply chain including: Logistics Planning, Order Expedition; Supply Planning; Global Demand-Supply Matching, Demand Planning, Inventory Planning and more.
Smarter Control Towers are Here Now
The autonomous level offers significant advantages – detecting more issues earlier, and resolving them faster, cheaper, and more optimally. Through increased automation, businesses can reduce the high manual overhead associated with many supply chain tasks, including order generation, inventory optimization, tracking, issue resolution, optimal demand-supply matching, and support. As a result, the business will be able to offer greater levels of service and value to end customers at the lowest possible cost.
Networks are an essential foundation for smarter control towers and make them possible. They connect and harmonize internal systems and external partners. This is key to making the transition from siloed legacy systems to a collaborative, real-time digital business network. They unify planning and execution to be more agile and intelligently coordinated. They can serve up data from across the network, through aggregation and filters that highlights the key information, KPIs and drive insights for higher level functions. And ultimately, the network-wide perspective provides better real-time information for human decision-making and for intelligent autonomous agents — so they can attain autonomy and begin executing resolutions where it makes sense.
About the Author
Joe Bellini is COO at One Network Enterprises, provider of an AI-enabled business network platform that enables all trading partners to manage, optimize and automate complex business processes in real-time. To learn more, visit www.onenetwork.com or follow One Network at https://www.linkedin.com/company/one-network-enterprises.