InterSystems Supply Chain Orchestrator

Data is the lifeblood of every supply chain organisation; as data grows, so does the prevalence of data silos. Organisations are striving to gain a competitive edge, deliver value to customers, reduce risk, respond more quickly to the needs of the business, and out-innovate the competition. Yet accessing, integrating, and transforming data from internal and external sources into actionable insights remains a challenge.

InterSystems Supply Chain Orchestrator™ is an AI-powered decision intelligence platform designed to enable the next generation of agentic and autonomous supply chain operations. It enables organisations to build a smart data fabric architecture that unifies data access, business context, advanced analytics, workflow orchestration, and AI capabilities, transforming fragmented supply chain data into actionable decisions and executable actions.

A key differentiator, and benefit, of Supply Chain Orchestrator is that it provides all of the key data management capabilities that are required — in a single, extensible platform, built from the ground up on a single architecture. This eliminates the need to implement, configure, and integrate multiple different data management services.

Supply Chain Orchestrator provides a comprehensive framework which includes the following built-in supply chain accelerators that speed and simplify the development of custom applications:

• Extensible supply chain data model
• Built-in supply chain analytics cubes
• Key performance indicator (KPI) framework
• Automated issue detection
• Issue lifecycle management
• Advanced analytics for issue resolution
• Supply chain APIs
• Scenario analysis
• AI assistant with supply chain agents
• Data engine that supports a wide range of data types, including multi-model, multi-access, vector, structured, and unstructured data, all in one engine
• Data integration and interoperability
• Business intelligence
• Business process and rules engine

Despite commonalities between different supply chains, each supply chain is unique in some way, thus making every business’ use case unique. There can never be a “one-size fits all” data model that will meet the needs of all types of supply chains. Therefore, it is important that the canonical supply chain data model can be extended or customised to meet the needs of each specific use case. The data model provided by Supply Chain Orchestrator supports the following features:

• Custom data attributes: Custom data attributes can be added to any existing data objects in the model. This can be done by customising a data object class or simply through an API call.
• Custom data objects: If the canonical data model does not provide the business entity required, a new object can be created. A new data object can be created through a new class definition, or simply by making an API call.
• API support for data model extensions: As mentioned above, APIs can be used to add custom attributes or to create new custom objects. APIs are also provided to inspect the current data model details: what objects are system defined and which ones are custom; what attributes are defined for each object; and which ones are custom attributes.
• API support for data access: Once the data model is extended, APIs can be used to access the extension by adding data to the extended data model, query data based on custom attributes or on custom objects, or other CRUD (create, read, update, and delete) operations on the data anywhere in the extended data model.
• Upgrade safe: Any extension of the data model will be preserved during Supply Chain Orchestrator upgrades, even if the canonical data model is enhanced in a new release. No additional data migration step is required for upgrades.

With embedded analytics and a canonical supply chain data model, clients can get immediate value from their supply chain data once loaded, such as business intelligence dashboards or reports. With a smart data fabric architecture, there is no need to move the data from a transactional schema to an analytics schema, and analytics cubes can be defined directly on the supply chain data model. To speed this up even further, data cubes for key supply chain data objects, such as orders, shipments, inventories, and issues are prebuilt frameworks within Supply Chain Orchestrator. These cubes can be used for configuring custom dashboards, generating business intelligence (BI) reports, or used by other supported BI tools. The data cubes provided out of the box can be extended with new measures or dimensions, based on data model extension or customisation. But new cubes can also be configured independently. In addition to the common BI and reporting usages, cubes are also used by Supply Chain Orchestrator as the foundation for its KPI framework.

Many supply chain activities are driven by KPIs, which can be used for many different purposes such as tracking business objectives and goals or detecting risks in a supply chain. Conceptually, there are many common supply chain KPIs, such as on-time in-full (OTIF) orders or aging inventory, but the actual logic behind these KPIs can still vary from business to business. Supply Chain Orchestrator offers a KPI framework that can be used to configure KPIs based on a client’s specific logic. The KPI framework allows clients to define KPIs with the following details:

• KPI logic: For example, the logic for the definition of “late” for a late ship order KPI.
• KPI dimensions: If a client is interested in learning late ship orders by country and by products, the KPI definition can include country and product as KPI dimensions.
• KPI thresholds: Two KPI threshold values can be defined for each KPI, watching threshold and warning threshold.
• KPI value type: Two types of values can be used for a KPI, raw value (such as count of late orders, or sales revenue dollar amount), or a percentage value (percentages of orders that are late).
• Issue flag: A KPI can be used to autogenerate issues for any data records to meet the KPI condition. For example, one can make late ship order KPI to be an issue-generating KPI, so any orders shipped late will have an issue generated and tracked in the system.

Supply chain disruptions and risks are modeled as issues in Supply Chain Orchestrator. Issues can be automatically generated based on KPIs, triggered from business processes, or imported from external systems. Once an issue is saved in Supply Chain Orchestrator, its lifecycle can be managed within the system, including setting issue status, running issue analysis, providing actionable insights, etc. Supply Chain Orchestrator also provides out-of-the-box issue analysis to help clients understand issues in different categories, business impacts by different type of issues, and statistics on issue status.

A key part of issue lifecycle management is issue analytics, which can provide insights to the following about an issue:

• Severity level: How big an impact is it for the business?
• Urgency level: How time-critical is the issue? Root cause analysis, e.g., what triggered the problem?
• Impact analysis: What would the impact of this issue be if it is not properly addressed?
• Prescriptive actionable insights analysis: What are the recommended actions to mitigate the risk of the issue and related business impacts?

Although the analysis of each business challenge is different, Supply Chain Orchestrator provides the key infrastructure and framework to simplify the development or configuration of related business processes and business rules for the above analysis needs, making it ideal for both systems integrators and application software developers.

APIs are provided for accessing all Supply Chain Orchestrator features, including:

• Data model APIs for model discovery and model extension.
• Data access APIs for data management including create, read, update, and delete (CRUD) operations on any supply chain data and search capabilities. All data access APIs support user defined pagination and sorting, which simplifies the work of related UI developments.
• KPI-related APIs, including listing defined KPIs, creating new KPIs, getting a KPI value or values, getting a list of data records associated with a KPI, etc.
• An API for issue management, including new issue creation, search/retrieve issue details with analysis results, run issue analysis, and closing issues.

In addition to the APIs listed above, Supply Chain Orchestrator also provides many other APIs for different aspects of the data platform.

Unlike AI solutions that operate as standalone copilots or isolated AI assistants, Supply Chain Orchestrator embeds AI directly into operational decision workflows, enabling organizations to progress from visibility to intelligence, and ultimately to autonomous execution.

InterSystems believes that successful agentic supply chains require four foundational capabilities:

• An intelligent and contextualised data foundation.
• Decision intelligence that combines analytics with business knowledge and operational context.
• AI agents that augment human decision making and automate routine activities.
• Workflow orchestration and governance that enable decisions to become actions.

Supply Chain Orchestrator combines these capabilities within a single architecture.

Effective AI begins with trusted, contextualised, and accessible data. Supply Chain Orchestrator provides a unified data foundation that supports transactional, analytical, operational, and AI workloads on a single platform. This intelligent data foundation eliminates data silos while providing the business context required for reliable AI-driven decisions.

Supply Chain Orchestrator combines enterprise data, operational context, business policies, and supply chain expertise to support intelligent decision-making. The objective is not simply to answer questions, but to help organisations understand options, evaluate trade-offs, and determine the best course of action. This enables organisations to move beyond dashboards and reporting toward actionable intelligence and operational decision support.

Supply Chain Orchestrator provides specialised AI agents that augment human decision-making and support intelligent business processes.

• Data agents allow both technical and business users to discover, access, prepare, enrich, and contextualise information from across the enterprise.
• Business process agents monitor operations, identify exceptions, coordinate workflows, and recommend or execute actions according to customer-specific business policies, rules, and operational processes.
• Memory and context agents maintain business context, organisational knowledge, historical decisions, customer-specific operating practices, and industry expertise.

Organisations adopt AI at different speeds and with different governance requirements. Supply Chain Orchestrator supports multiple levels of operational autonomy:

• Human-in-the-loop decision support, where AI agents analyse situations, generate recommendations, and present options for user approval.
• Human-on-the-loop automation, where users provide oversight while agents execute approved workflows and manage routine decisions.
• Fully autonomous execution, where AI agents monitor events, make decisions, and trigger actions automatically within predefined policies and governance boundaries.

This approach allows organisations to increase automation progressively while maintaining operational control and regulatory compliance.

Supply Chain Orchestrator™ AI Assistant provides natural language support and allows business users to interact with features supported in Supply Chain Orchestrator. For example, a user can query any supply chain data in Supply Chain Orchestrator using their own language/terminology or ask for an explanation of the recommendation that Supply Chain Orchestrator is making.

AI Assistant focuses on the following capabilities:

• Accessing supply chain data: This can be a direct query of a specific data entry (like a customer order), or a query on aggregated data (such as revenue by region and timeline).
• Decision assistant: Helps users understand the business logic/rules used for actionable insights through Supply Chain Orchestrator issue analysis.
• Memory management: There are many reasons memory is important for user experience with AI solutions, such as to help AI understand a specific industry better, to minimise user prompts by keeping user specific preferences or settings etc.

Supply Chain Orchestrator is a powerful decision intelligence platform for any organisation in any industry. It allows application developers to bring together capabilities from different applications through configuration and composition.

As supply chain organisations are faced with the need to make real-time decisions amidst increasing amounts of data, the challenge lies in breaking down data silos, providing end-to-end visibility, and real-time embedded predictive and prescriptive analytics to respond more quickly and accurately to exceptions and disruptions. With a focus on real-time data sharing and trusted insights, Supply Chain Orchestrator empowers organisations to optimise their supply chain operations and enhance overall efficiency.

By seamlessly integrating disparate data sources, Supply Chain Orchestrator empowers decision-makers, driving better outcomes and delivering value to customers. For those navigating today’s increased complexities of supply chain management, Supply Chain Orchestrator provides one reality of the supply chain, powered by unified data.

1 - Optimising Supply Chain Performance with Unified Data