Tool enables business analysts to create analytics-powered decision services, “point and click” without coding, using optimization, Monte Carlo simulation, data mining and machine learning, rules and decision tables; works in Excel for the Web, Windows and Macintosh and with Microsoft Teams.
Frontline Systems is shipping Analytic Solver V2022, a new version of its advanced analytics toolset for Excel (Web, Windows and Macintosh), that enables business analysts to easily build models using business rules, machine learning, mathematical optimization and Monte Carlo simulation, and easily deploy those models in cloud-based applications.
Analytic Solver is upward compatible from the Solver in Excel, which Frontline originally developed for Microsoft, and is able to solve virtually any type or size of optimization model – using methods from linear programming to genetic algorithms and stochastic optimization, ranging from a few to millions of inter-related decisions in a single model.
“With Analytic Solver, business analysts have powerful optimization, risk analysis and machine learning capabilities – and they can ‘move beyond Excel’ to gain cloud-based data access, model management and governance, without losing the advantages of Excel,” said Daniel Fylstra, Frontline’s President and CEO.
Top iTechnology Security News: Q-Net Security adds Cybersecurity Leaders to its Board of Advisors
Decision Intelligence Results in Months, not Years
Analytic Solver works with RASON®, Frontline’s Azure-hosted platform, to empower users to ‘publish’ and manage analytic models as RESTful decision services. The combination is a complete “decision intelligence suite” that supports business rules, forecasting, machine learning, optimization and simulation methods, from small models to large, multi-stage analytics workflows.
With Analytic Solver, users don’t need coding skills, ‘DevOps’, ‘ModelOps’ or IT expertise – they use a Deployment Wizard to turn their Excel models into a form easily usable in Microsoft Teams, Power BI, Power Apps, Power Automate, or any application that can consume JSON or OData. These models can be used directly for classification and prediction (no auxiliary code such as R or Python needed), and they are easily managed, governed, and even run in multi-stage ‘decision flows’ with RASON’s facilities – gaining the benefits of Azure authentication and cybersecurity.
Users can also deploy and share probability models, following the open Probability Management 3.0 standard – a major benefit for large organizations with multiple risk analysis and modeling projects.
Higher Performance Optimization
Besides including the latest, fastest-ever versions of optimization solvers such as the Gurobi Solver 9.5 for linear, quadratic and mixed-integer problems, Analytic Solver V2022 offers fastest-ever performance in ‘model setup’ – the process of analyzing a model with complex, inter-linked formulas and use of Excel’s many built-in functions, and converting this into the relatively simple form that optimization solvers can use. For larger models, this can cut end-to-end solution time by 50% or more.
These performance enhancements apply to both Analytic Solver Desktop, which solves models on users’ own PCs, and to Analytic Solver Cloud, which solves models “in the cloud” via the RASON service. RASON in V2022 includes further enhancements to its ability to concurrently solve a wide range of models from many users and different sources.
Top iTechnology Cloud News: phoenixNAP and CloudGenera Collaborate to Provide Proof of Value for Cloud Infrastructure Use
Automated Machine Learning with Find Best Model
Analytic Solver offers “augmented machine learning” features found only in other sophisticated machine learning tools. The user simply supplies data in tabular form: Analytic Solver will test and evaluate multiple types of machine learning models – classification and regression trees, neural networks, linear and logistic regression, discriminant analysis, naïve Bayes, k-nearest neighbors and more, validate and compare them according to user-chosen criteria, and deliver the model that best fits the data.
Better Simulation Models with Metalog Distributions and Fitting
In risk analysis and Monte Carlo simulation, Analytic Solver and RASON offer leading support for the increasingly popular Metalog family of distributions, created by Dr. Tom Keelin, that can closely approximate virtually any classical continuous distribution, and better fit historical data than classical distributions. Only Analytic Solver will automatically fit user data to the full range of possible (bounded and unbounded, multi-term) Metalog distributions.
More Optimization, Simulation and Data Science Enhancements
Analytic Solver users receive major functional and performance enhancements for analytic modeling every year. Besides new features in this release, enhancements in the last 18 months include:
- 15 new PSI Simulation Functions enable easy cloud use of time series simulation methods
- 9 new PSI Optimization Functions enable live sensitivity results on worksheets
- Use custom functions created with Excel’s LAMBDA or DMN’s FEEL features in models
- Multi-stage “data science flows” in Analytic Solver are deployed as RASON decision flows
Top iTechnology Security News: Creating a Contactless Travel Experience: Ottonomy Launches First Fully Autonomous Delivery Robots at CVG Airport
[To share your insights with us, please write to sghosh@martechseries.com]
