We at Sparkling Logic help businesses automate and improve their operational decisions through our revolutionary decision management technology. In this post, we present our SMARTS™ decision management platform through 10 common questions and answers.
1) What is SMARTS™?
SMARTS™ is a cloud-first, enterprise-level business rules and decision management platform. Our technology enables non-technical business analysts and domain experts to leverage their own data to easily design, test, deploy, monitor, and continuously improve business-critical decisions. Whether these decisions are explicit or AI-driven, users can effectively manage them with minimal IT resources.
As a result, our clients are able to reduce operational costs while delivering a better customer experience, modify decisions and launch new products faster, improve the quality and consistency of decisions, increase flexibility, scalability, and transparency, and effectively operationalize AI, machine learning, and deep learning models in decision-making.
SMARTS™ is available as a zero-install web-based user interface for configuration and a standards-compliant service interface for integration into applications as well as the wider enterprise eco-system (authentication providers, enterprise notification systems, etc.). SMARTS™ can be hosted as PaaS, hosted on customer cloud subscriptions and deployed as a service, or embedded as a component in applications.
2) Is SMARTS™ a Business Rules Engine?
SMARTS™ is much more than a business rules engine. It is a decision engine that integrates multiple decision technologies into the same platform. SMARTS™ supports various decision logic representations including business rules that users can easily combine and implement. In addition, SMARTS™ has multiple execution engines dedicated to optimizing the various decision logic representations in runtime. For example, our compiled Rete-NT inference rules engine is ideal for rules with many interdependencies. Beyond authoring and deploying decision logic, SMARTS™ provides various tools and capabilities for complete lifecycle management including access control, testing, experimenting, monitoring, versioning, and much more.
3) What are Common Use Cases for SMARTS™?
In financial services, insurance, and healthcare, our clients choose SMARTS™ to manage their decision logic for account management, fraud detection, credit origination, underwriting, pricing, rating, and collections. In general, SMARTS™ is a good fit when there are a lot of decisions that are data-based, frequently invoked, and likely to change.
4) Why Use SMARTS™ When You Can Code Decision Logic Directly into Your System?
The challenges with managing code is why decision management technologies like SMARTS™ were developed in the first place. In cases where you decision logic is simple and not likely to change, manual code works just fine. However, when you’re dealing with complex decision logic that is likely to change and is used through multiple applications, you’ll have a much easier time managing that logic in SMARTS™. With code, the onus is on IT to manage that logic. However, this is not ideal when the decisions are domain-specific, such as underwriting in insurance. SMARTS™ empowers domain experts to manage the decision logic themselves and to deploy that logic on its own lifecycle outside the software development lifecycle. This enables businesses to adapt faster to regulatory and other market changes.
5) Does SMARTS™ Include a Decision Design Process?
Yes! In addition to tools for authoring decision flows and business rules, SMARTS™ includes a tool for creating DMN diagrams. From the Object Management Group (OMG), Decision Model and Notation (DMN) is a standard for modeling an organization’s decisions. SMARTS™ Pencil is a graphical tool that enables users to draw DMN diagrams to capture knowledge sources, data inputs, and decision outputs.
6) What Machine Learning Models Do SMARTS™ Support?
SMARTS™ has over 17 predictive model execution engines to optimize machine learning model runtime performance. Common techniques include regression (linear, polynomial, logistic), neural network, decision tree, clustering, and classification (Naive Bayes, support vector machines, k-nearest neighbors, random forests). Users can upload any model in PMML format and execute it from SMARTS™. However, there may be situations where a model must run as a separate service. SMARTS™ also provides support for remote functions so that users can invoke a model through JSON-RPC or REST.
7) Does SMARTS™ Integrate with Business Process Management Systems (BPMS)?
Yes! A BPM system or platform can invoke a SMARTS™ decision service like any other service. In fact, a SMARTS™ decision service can be invoked by any system or platform. For example, in financial services, your loan origination system can invoke SMARTS™ at various decision points within a loan cycle, such as fraud detection or credit scoring. Or in insurance, your policy administration system can invoke SMARTS™ for underwriting decisions. In addition, SMARTS™ has an Orchestrations tool that users can leverage to automate decision management workflows such as retrieving, manipulating, and storing data.
8) What is the Difference Between an RPA Tool and SMARTS™?
A process is a series of decisions and actions. SMARTS™ is used to define, automate, and manage the decisions (specifically the operational ones) within a process. RPA tools, on the other hand, are used to automate specific actions within a process. Therefore, an organization can use both SMARTS™ and various RPA tools (integrated with or without a BPMS) to automate and/or optimize a process.
9) Is SMARTS™ cloud-based?
Yes! We designed SMARTS™ from the ground-up for the cloud. Whether you choose to host your application or use our SaaS solution, we provide you with the most modern tools. SMARTS™ comes in a container, ready to install on your premises, AWS, GCP, Azure, or Aliyun. You can change your hosting method without needing to recode to redeploy your application.
10) What makes you unique?
- Data and Performance Centric: SMARTS™ enables users to develop, maintain, and optimize decisions in the context of their data and decision analytics. This continuous feedback ensures decision consistency and quality throughout their lifecycle. Users can create and run Regression Tests, Simulations, and Champion/Challenger Experiments by defining QA flags, KPIs, and other metrics and view results in our customizable Dashboard. Users can also perform execution traces and data-aware, step-by-step debugging for further analysis of failed test cases. Our real-time metrics enables users to monitor live performance including any deployed champion-challenger and shadow experiments.
- Modern Architecture: SMARTS™ is designed for speed, scalability, and high availability. We leverage the industry’s fastest execution algorithms and our microservices approach allows each service to scale horizontally and vertically. Our deployments leverage cloud standard orchestrations such as Kubernetes/Helm and service meshes such as lstio. We support multi-site, multi-cluster installations with redundancy at development, testing, and production time and also support auto-scaling as required.
- Generative AI Support: AI Assistant is our generative AI-powered tool embedded in SMARTS™ that provides guided, on-demand support throughout the decision management lifecycle. Users can chat and interact with AI Assistant to accomplish decision management tasks such as authoring and modifying business rules using natural language.
Learn more about Sparkling Logic’s SMARTS™ Data-Powered Decision Manager