Program

(Click on the presenters or scroll down to see presentation abstracts and author biographies)

Presentation Abstracts and Author Biographies

No Code Decision Model using ChatGPT by Gary Hallmark (Oracle)

Natural language is the new no code development paradigm. We use the public and free (as of May 2023) version of ChatGPT to generate all of the decision logic in the well-known DMN Loan Approval Strategy decision using simple English prompts and natural language decision tables. The generated code often includes unit tests without even asking for them. We can assemble the generated logic into a decision service, without coding, using a simple ChatGPT prompt. Even the decision requirements diagram above is generated from a natural language prompt. The generated code can be assembled in a Jupyter notebook and deployed as a serverless function in a cloud such as Oracle or AWS.

After a demonstration, I’d like to engage in an open discussion about the impact of large language models, including
– Do we finally have a friendly enough language?
– Is the point and click approach to no code over?
– Can natural language handle very large models?
– Should LLMs generate code to make decisions, or make the decisions directly?
– What if decisions are based on sensitive information?
– Will AI services charge more for generating code? Will the code be licensed?
Keywords: chatgpt, LLM, DMN, no code, natural language

Gary Hallmark has been an Architect at Oracle for over 25 years. He has worked on databases, business rules and processes, standards, low-code user interfaces, application integration, chatbots, and cloud architecture. Gary participated in four iterations of the OMG Decision Model and Notation (DMN) standard, and was instrumental in the creation of FEEL and boxed expressions for decision logic.

Low Code Neuro-Symbolic Agents: Prompt Engineering through Process and Decision Orchestration by Denis Gagne (Trisotech)

In this presentation, we delve into the exciting world of Neuro-Symbolic AI and explore how process and decision orchestration of prompts can enable the powerful technique of prompt engineering. We begin with an introduction to Language Models as Systems (LLMS) and their capabilities, setting the stage for understanding the significance of prompt engineering.
Prompt engineering has emerged as a key methodology in leveraging the capabilities of language models to obtain desired outputs by providing explicit instructions or constraints. By carefully crafting prompts, we can shape the behavior of language models and align their outputs with our intentions. This presentation explores how the fusion of symbolic AI and prompt engineering can enhance the reliability, interpretability, and performance of these new forms of AI systems.
Next, we delve into the building blocks of prompt orchestration, exploring the techniques and strategies involved in designing effective prompts. We explore the concept of prompt engineering as a multidimensional task that involves a combination of linguistic framing, reasoning, and constraint setting. We discuss how symbolic reasoning and neural networks can collaborate in a synergistic manner to augment the capabilities of AI systems. Of particular interest is the creation of generative agents leveraging process and decision orchestration of prompts.
To bring the concepts to life, we present a live demonstration showcasing the applications of this novel neuro-symbolic prompt engineering. Through a series of interactive examples, we highlight how prompt orchestration can be utilized to achieve specific outcomes, while ensuring transparency and interpretability.
In summary, this presentation aims to provide a comprehensive understanding of a form of neuro-symbolic AI, prompt engineering, and the role of process and decision orchestration in achieving optimal outcomes. By exploring the building blocks of prompt orchestration and showcasing examples, we seek to inspire the audience to harness the power of prompt engineering and contribute to the development of responsible and effective AI systems. Keywords: Process and decision orchestration, Large Language Model systems (LLMs), Symbolic AI, Sub-symbolic AI, DMN, BPMN, CMMN, Agents

Denis Gagné is CEO and CTO of Trisotech, a leading Standard Based Low-Code Intelligent Automation enterprise software vendor. For more than two decades, Denis has been a driving force behind most international BPM standards in use today. Denis is a member of the steering committee of the BPM+ Health Community of Practice, where he also leads the Ambassador program. For the Object Management group (OMG), Denis is Chair of the BPMN Interchange Working Group (BPMN MIWG) and an active contributing member to the Business Process Model and Notation (BPMN), the Case Management Model and Notation (CMMN), and the Decision Model and Notation (DMN) work groups.

Human Values and the Business Intervention Model: Rediscovering Management Cybernetics in an AI-Sceptic World by Dr. Alan Fish (FICO)

The current wave of consternation around automated decision-making is exacerbated by the use of the term “Artificial Intelligence” with its implication of independently motivated non-human agents. A much more favourable mindset might have been fostered by the old term “Management Cybernetics”, describing a field which aimed to harness control and communication technology to serve human goals. Best practice in model-based decision management reflects the cybernetic approach by decomposing a domain of decision-making into a connected system of human and automated decisions, each of which can be optimised using feedback loops. However, we do not yet have a coherent approach for embedding human values as the goals of that system. I propose the Business Intervention Model (BIM): a conceptual framework for defining the business intent of decision management interventions and optimising decision-making over time. It uses Key Performance Indicators (KPIs), which traditionally measure business value for the enterprise, but can also express societal values by defining Environmental, Social and Governance (ESG) goals. To regain public trust we must adopt more appropriate and reassuring language, and justify this by redesigning our decision systems so that the values they embed are transparent and accessible. We must show that humans will remain at the centre of decision making, and that our decision systems exist to serve human needs and values. Keywords: ethical AI, human-centred AI, cybernetics, decision management, EDM, decision model, Business Intervention Model, BIM, KPI, ESG.

Dr. Alan Fish is a thought-leader in Decision Modelling and Decision Management. He invented the Decision Requirements Diagram (DRD) which exposes the structure of a domain of decision-making, and developed Decision Requirements Analysis (DRA): a methodology for building and using such decision models. Alan is the author of “Knowledge Automation: How To Implement Decision Management in Business Processes” (Wiley), which has been translated into Chinese. He is editor and co-author of the OMG specification Decision Model and Notation (DMN), and co-chairs the OMG DMN task force. He continues to develop notations, methodologies and ontologies to provide the conceptual environment for business users of FICO Platform.

Outside work Alan is a musician: singer-songwriter, guitarist and saxophonist. He is a keen member of several musical ensembles and has released an album of his own songs: Yes Why Not.

From Machine Learning integration to Decision Transparency by Guilhem Molines (IBM)

Decisions are more than rules. Slowly, Machine Learning models are being integrated into Decision Models to enrich them with predictive scores crafted by data scientists. But doing so introduces a black box in the middle of a model whose purpose was precisely to increase decision transparency. How can we model decisions benefiting from externally defined predictive models while still keeping the transparency feature? This session will explore solutions to this challenge. Keywords: decision, modeling, machine learning, transparency, explainability

Guilhem Molines, Decision Chief Architect, IBM. With a background in fundamental Computer Science and Artificial Intelligence, Guilhem Molines has been involved with decision technology for more than two decades, in various roles in the field and in the development Lab. Today, he is the Chief Architect of the IBM team building Decision Technology With a special focus on the knowledge modeling and business user experience, Guilhem is always in close contact with users and practitioners and willing to find innovative ways to make the authoring of decisions an easier task for the industry. Since 2020, Guilhem leads the architecture of the next generation of decisioning platform and has also been involved in the navigation system of the unmanned Mayflower Autonomous Ship.

Using DMN on Machine Learning Projects by James Taylor (Decision Management Solutions)

The use of Decision Model and Notation (DMN) models to describe decisioning solutions is well established. Many of these include Machine Learning elements – sub-decisions that are implemented by ML models. There is increasing interest in how to use DMN in projects that combine explicit business logic, machine learning and AI. There is also a sense that DMN models may allow for better specification and management of the ML development itself, allowing for a seamless transition to DMN when the deployment and operationalization (MLOps) phase is reached. This presentation will discuss some of the suggestions being made, experience using the notation in this context and progress on the DMN committee to support these projects.
Keywords: MLOps, ModelOps, ML, Machine Learning, AI, Artificial Intelligence, DMN, Decision Model and Notation, Business Rules, Decisioning

James is the founding partner of Decision Management Solutions. He is a leading expert in how to use decision modeling, business rules, and analytic technology to deliver Digital Decisioning. James is passionate about helping companies improve decision-making and effectively adopt advanced analytic technology. He provides strategic consulting to companies of all sizes, working with clients in all sectors to adopt decision making technology. James has spent the last 20 years working in decisioning and has led Decision Management efforts for leading companies in insurance, banking, health management, and telecommunications. James is the author of “Digital Decisioning: Using Decision Management to Deliver Business Impact from AI” (MK Press, 2019) and “Real-World Decision Modeling with DMN” (MK Press, 2016) with Jan Purchase. He previously wrote and Decision Management Systems: A practical guide to using business rules and predictive analytics, Process and Decision Modeling with BPMN/DMN with Tom Debevoise and Smart (Enough) Systems: How to Deliver Competitive Advantage by Automating Hidden Decisions (Prentice Hall) with Neil Raden and he has contributed chapters on Decision Management to multiple books, including Applying Real-World BPM in an SAP Environment, The Decision Model, The Business Rules Revolution: Doing Business The Right Way, and Business Intelligence Implementation: Issues and Perspectives, as well as many articles to magazines. He is a regular speaker, a Faculty Member of the International Institute for Analytics, and the author of a popular blog on Decision Management (JT on EDM).

Balancing Automated and Human Decisioning – One of the most prominent challenges of DMN by Stefaan Lambrecht (The TRIPOD for OPERATIONAL EXCELLENCE)

Decision-making is a complex and critical process that affects the success of businesses. Traditionally, decisions are made based on intuition, experience, and judgment, which can be time-consuming, inconsistent, and prone to errors. With the rise of technology, businesses are turning to decision modeling and automation systems to streamline the decision-making process. One such technology is the Decision Model and Notation (DMN), which provides a standard modeling language for decision logic.
Although automation can bring significant benefits to the decision-making process, it is not feasible nor desirable to eliminate human decisioning entirely. Human decisioning provides context, judgment, and empathy that cannot be replaced by automation. Therefore, DMN models do not only focus on automating decision logic, but also and equally important, on the interaction between automated and human decisioning.
The role of human decisioning in DMN The role of human decisioning in DMN models is to provide context, judgment, and empathy to the decision-making process. Human decisioning can be used to validate the results of automated decisioning, provide additional information, or override automated decisions when necessary. Human decisioning can also be used to identify and correct errors in decision logic, improving the accuracy and consistency of decision-making.
Combining Automated and Human Decisioning DMN models can be designed to incorporate both automated and human decisioning. This approach is known as the “human-in-the-loop” design. In this design, the DMN model is executed by a decision engine, but the model reaches out to human experts to validate the results or to complete the decisioning process. The human decision maker can provide additional information or context to the decision engine, improving the accuracy and consistency of decision-making and filling the gaps in automated decision-making.
The human-in-the-loop design can also be used to improve the performance of the decision engine over time. By collecting feedback from human decision makers, the decision engine can learn from its mistakes and improve its decision logic. This approach is known as “machine learning” and can lead to more accurate and efficient decision-making.
How to Include Human Decisioning in DMN? Using a real-life business case I will demonstrate how to set up a DMN decision model in which the DMN model orchestrates automated and human decisioning, in order to keep full traceability of the end-to-end decision-making process. The content of the business case is based on a real-life scenario around compliance controls in a financial institution.
Keywords: DMN, Automated Decisioning, Human Decisioning, Straight-Through Processing, Traceability, Decision Modeling, BPMN and DMN Interaction

As a very experienced business architect and a business process guru, Stefaan brings companies to the next level of customer experience and operational excellence. With an educational background as interpreter and political scientist, Stefaan has always been a visionary outsider in the twilight zone between business and ICT. From the early 90s onwards, Stefaan understood the power of business processes in achieving customer-oriented operational excellence and in aligning business with IT. Specialties: Strategic Performance Management, Enterprise Architecture, Business Process Analysis & Redesign, Business Rule/Decision Management

GPT-3 for Decision Requirements Modeling and Advice by Jan Vanthienen and Alexandre Goossens (KU Leuven)
Decision models contribute significantly to the understanding of the requirements and the logic of operational business decisions and processes. The Decision Model and Notation (DMN) standard allows automating these decisions. This presentation examines if GPT-3 can be leveraged to automatically discover the decision requirements described in a textual description. The performance of GPT-3 on this task is compared to a pipeline making use of classifiers and a deep learning BERT model finetuned for Named Entity Recognition to extract decision requirements from text. Even though GPT-3 can analyze decision descriptions, it is not always capable to correctly execute or explain a decision. This session also illustrates how an automatic chatbot for DMN models enhanced with the natural language processing capabilities of GPT-3 leads to explainable decisions which can be interacted with in an intuitive manner. This enhanced solution offers stakeholders more transparency and trust in the decision process, without expensive and ineffective customer support. Keywords: Decision Modeling, DMN, GPT-3, Large Language Model, Chatbot

Prof. Jan Vanthienen received his PhD degree in Applied Economics from KU Leuven, Belgium. He is a full professor of Information Systems at the Department of Decision Sciences and Information Management, KU Leuven and (co-)authored more than 200 full papers in international journals and conference proceedings. His research interests include information and knowledge management, business rules, decisions and processes, and business analysis and analytics. He received an IBM Faculty Award on smart decisions, and the Belgian Francqui Chair at FUNDP. Currently he is department chair at the Department of Decision Sciences and Information Management of KU Leuven.

Alexandre Goossens is a third year PhD student in Applied Economics from KU Leuven, Belgium within the research team of Prof. Jan Vanthienen. His research spans multiple domains with a general theme of decision modeling, execution and discovery. More specifically, his research deals with extracting DMN models from textual descriptions using NLP and deep learning as well as mining DMN models from object-centric process event logs. Lastly, his research also deals with automated DMN chatbots which perfectly fit in the current trend of explainable AI.

Declarative Decision Modeling with Rule Solver by Dr. Jacob Feldman (OpenRules)
I will present a new approach that supports declarative decision modeling by integrating rule engines and constraint solvers within the same decision models. It allows business analysts to create a decision model using DMN-like decision tables (“WHAT”) and rely on predefined search strategies to come up with the proper decisions (“HOW”). The proposed approach introduces new types of decision table columns that define constrained variables and their relationships allowing a user to mix and match business rules with technical constraints within the same decision tables. I will use well-known decision modeling problems to compare this new approach to traditional procedural decision modeling or pure programming approaches.
Keywords: Declarative Decision Modeling, Rule Engine, Constraint Solver, Smart Decision Engine, Rule Solver

Dr. Jacob Feldman is the CTO of OpenRules, Inc., a US corporation that created and maintains the highly popular Business Rules and Decision Management System commonly known as “OpenRules”. He has extensive experience in development of decision-making engines using business rules, optimization, and machine learning technologies for real-world mission-critical applications. Jacob is the DecisionCAMP’s Chair,  the manager of DMCommunity.org, and an active contributor to BR&DM forums. He is also the Specification Lead for the optimization standard JSR-331. Dr. Feldman is an author of two books “DMN in Action with OpenRules“ and “Goal-Oriented Approach to Decision Modeling“. He has 5 patents and many publications in the decision intelligence domain.

Rules: Shaping Behavior and Knowledge by Ronald G. Ross (Business Rule Solutions)

Explicit guidance for business and government requires a rich cast of characters. Playing the central role, of course, is rules. But rules are not just for inference and decisions, they are also for definitional and behavioral matters. The latter involves real-time insertion of sentiment, human discretion, and common sense that are so often lacking in today’s digital systems. Beyond the main actors, the supporting cast is broad and rich. It includes rights, warranties, permissions, authorizations and more – all in a unified base that is fully explainable and traceable. No black boxes!

The Holy Grail of rules? This presentation presents it as a comprehensive set of guidance components for business and government in natural language that is directly computable.
• Rules for groups and communities of people
• Guidance bases that can be automatically checked for consistency prior to activation
• Support for a broader base of problems, including regtech, lawtech, fintech – and you-name-it.

Keywords: rules, guidance, behavioral rules, definitional rules, rights, warranties, permissions, authorizations, governance

Ronald G. Ross is Co-Founder and Principal of Business Rule Solutions, LLC (www.BRSolutions.com). BRS provides consulting, training and mentoring in support of policy analysis, business rules, concept modeling, decision analysis, and business knowledge engineering. BRS clients have included many 100s of top businesses and government bodies world-wide. Ron is the author of the 2020 groundbreaking book “Business Knowledge Blueprints: Enabling Your Data to Speak the Language of the Business”, featuring concept models, business vocabularies and disambiguation. It is his 9th professional book.

Ron is Chair of the annual Building Business Capability (BBC), official conference of the IIBA®. He is also Executive Editor of BRCommunity.com and its flagship on-line publication, Business Rules Journal. Ron has keynoted dozens of conferences and given seminars to many thousands of people worldwide. Ron co-develops the landmark BRS methodology featuring numerous innovative techniques including the popular RuleSpeak® (free on RuleSpeak.com). These are the latest offerings in a 45-year career that has consistently featured creative, business-driven solutions. Ron is recognized internationally as the ‘father of business rules.’ In 2017 he was co-author with John Zachman and Roger Burlton of the Business Agility Manifesto (www.busagilitymanifesto.org). Mr. Ross holds an M.S. in information science from the Illinois Institute of Technology and a B.A. from Rice University.

Powering Risk Mitigation Analytics in Energy Infrastructure—Case Study and Technical Approaches by Seth Meldon (Progress)

Maintaining and optimizing municipal energy infrastructure—everything involved with the transmission and distribution of electricity from centralized power plants—is an exercise in holistic prioritization across a considerable breadth of internal and external demands. To adhere to these multi-dimensional, fast evolving requirements, a power utility company serving more than one million customers sought to innovate in a way that maximized and capitalized upon the expertise of its skilled workforce of electric engineers, not to supplant this workforce by automating their responsibilities.

With a rules-based risk calculation system comprised of rules authored in Progress Corticon by experienced industry practitioners, the organization is building a novel solution to a set of requirements with mounting importance, including the decarbonization of energy sources, planning for and mitigating risk from the increasing frequency and severity of environmental disasters spurred by global heating, and maximizing the technical efficiency with which electricity is transmitted across long distances. The rule-driven aspects of these business requirements which are certain to expand in their prevalence, standardization, and regulatory focus worldwide will be the focus of this session.

Seth Meldon works as a senior solution engineer at Progress Software, where for more than five years he has had a product focus on the Corticon Business Rules engine. He serves as a primary technical resource for prospective and current customers, while collaborating with colleagues across the organization to build innovative industry solutions from Progress’s suite of application development products.

Generative AI and Regulatory Compliance by Denis Gagné (Trisotech) and Tom DeBevoise (Advanced Component Research)

Meeting regulatory compliance can be a daunting task for all businesses, particularly in the banking and finance industry. To comply with regulations, an organization must align its operations with the specified obligations. This involves integrating the regulations’ business rules into the decision models. However, proving that a decision aligns with regulatory requirements can be challenging since the implementation often involves a subjective interpretation of the regulations. Moreover, many organizations rely on informal or inconsistent approaches to extract decision requirements from the regulations. Generative AI can assist with regulatory compliance in several areas: • Extraction of important terms and business rules from the regulations • Descriptions of terms and concepts according to the regulations • Code generation, in FEEL, for a subset of regulations To use Generative AI assistance, we utilize a reliable approach to comprehend the decision requirements of the regulation fully. We propose creating a knowledge entity model (KEM), which is a conceptual model that accurately represents the operations of a regulated business area. This is achieved by creating a vocabulary of precise terms, linking them through concept maps, and integrating business rules that connect concepts, terms, and operational limitations. All these elements are logically linked to the regulations, allowing crucial areas of the regulations to be viewed as business rules. The KEM is built on the OMG standard Semantics of Business Vocabulary and Business Rules (SBVR). Using a KEM to model regulations, in conjunction with generated FEEL code from Generative AI, permits a traceable implementation. Using prompt engineering to extract the key terms and concepts from the regulations accelerates the implementation, reduces errors, and minimizes the need for subjective interpretations. Keywords: Regulation, Laws, Compliance, Generative AI, DMN, FEEL, SBVR

Denis Gagné is CEO and CTO of Trisotech, a leading Standard Based Low-Code Intelligent Automation enterprise software vendor. For more than two decades, Denis has been a driving force behind most international BPM standards in use today. Denis is a member of the steering committee of the BPM+ Health Community of Practice, where he also leads the Ambassador program. For the Object Management group (OMG), Denis is Chair of the BPMN Interchange Working Group (BPMN MIWG) and an active contributing member to the Business Process Model and Notation (BPMN), the Case Management Model and Notation (CMMN), and the Decision Model and Notation (DMN) work groups.

Tom Debevoise has extensive experience in Process and decision modeling using BPMN, DMN and FEEL. Tom Debevoise focuses on next-generation IT solutions for business operations as a technology leader and cloud solutions architect. Tom works on the next generation of intelligent, practical, cloud-based services. Tom is developing these with an “intelligent digital assistant”, using Natural Language Processing, APIs to massively integrated services, and a core set of responsive processes, decision making, and analytics. Tom has held various positions at companies such as Oracle, Bosch, and Signavio.

DMN On-Ramp: Defining Business Use Cases for DMN and their Tool Requirements by Jan Purchase (Lux Magi) and Ryan Trollip (Decision Management Solutions)

How well does your DMN tool support your business use case? What functionality might you need at the next stage of your adoption? Join us and find out. The current marketplace can confuse businesses wishing to benefit from digital decisions’ transparency, agility and powerful AI integration capabilities. Although DMN defines an invaluable notation for representing decisions, the DMN standard documentation is large, complex, and not aimed at end-users. Furthermore, although it describes the details of the DMN exhaustively, it does not (by design) include any business case for using DMN or any means of assessing what tools are necessary to support it. Furthermore, the standard presents vendors with an all-or-nothing approach to conformance, leading to large-scale inconsistency in the tools market. Launched at DecisionCAMP 2021, The DMN On-Ramp aims to supply a vendor-independent, validated, staged route for DMN adoption and tool conformance aimed at newcomers trying to establish a decision management capability and DMN tool vendors. The committee’s mix of practitioners, vendors and academics has clarified the different ways businesses can benefit from DMN and the functionality set of tools required to support these use cases. An interactive website explaining both will soon be released. In this presentation, we cover some of the newer use cases for DMN, including integration of AI techniques such as machine learning, identifying data requirements and tracking the evolution of rapidly evolving decision logic. In addition, we discuss the tool requirements necessitated by these use cases and how they might be fulfilled. Attendees of this session will gain an understanding of the many use cases for DMN and the new demands these applications place on DMN tools.
Keywords: Business adoption of DMN, DMN Tool Requirements, Decision Intelligence, Interpretability

Dr Jan Purchase has been working in investment banking for 25 years, during which he has worked with nine of the world’s top 40 banks by market capitalization. Over the last 20 years he has focused exclusively on helping clients with automated business decisions, decision modelling and integration of artificial intelligence (including machine learning) into business decisions. He is a founder of Lux Magi (www.luxmagi.com): a company specializing in delivering, training and mentoring all of these concepts to financial organizations. Lux Magi has been applying decision modelling to the automation of finance and regulatory compliance since 2011 to support their clients in becoming and remaining compliant with ever changing regulations. In 2021, Lux Magi was one of the first companies to pioneer the integration of large language models and decision modelling for corporate fact checking.
Jan has maintained a blog ‘Decision Management for Finance’ since 2010, highlighting the practical lessons learned from applying decision modeling and machine learning at scale and providing useful feedback to clients and vendor partners. He has published many white papers, hosted multiple webinars and chaired public coaching sessions on the application of decision modeling and artificial intelligence to problems in finance. Dr Purchase is co-author (with James Taylor) of the book Real World Decision Modeling with DMN a comprehensive guide to decision modelling with DMN of which the second edition (for DMN 1.5) is due imminently

Ryan Trollip is CTO and Vice President of Services of Decision Management Solutions and the founder of DecisionAutomation.ORG. He is an ILOG alumnus with over 20 years in consulting leadership experience across South Africa, the UK and the US. Over the last 10 years, Ryan has built and led decision management and enterprise architecture practices worldwide and successfully delivered some of the largest and most complex decision management solutions in leading Fortune 500 companies; implementations including Next-Best-Action, omni-channel systems for multiple Financial Institutions as well as helping MasterCard develop and roll out their industry-leading fraud detection solution. He is passionate about providing consistent, quality solutions that leverage actionable business insights to provide business value as well as clear ROI. Most recently Ryan led a team of enterprise architects helping customers achieve business and technical transformations. By leveraging industry methodologies to create blueprints and road-maps he assisted clients with planning and execution of strategic initiatives. Ryan has spoken with customers at IBM’s Impact and InterConnect conferences, the Building Business Capability co

User-friendly Probabilistic Decision Logic Modeling by Simon Vandevelde (KU Leuven)

The goal of this presentation is to highlight the synergy between DMN and probabilistic reasoning, and to present our probabilistic DMN extension called pDMN. In real life, uncertainties are found almost everywhere; yet, DMN has no native support for probabilities. While DMN’s preciseness and determinism are one of the main advantages of the notation, there are cases in which it is still useful to include uncertainty. To this end, we present the pDMN notation, which extends DMN with probabilistic reasoning while still maintaining all of the usual decision table features.
Keywords: DMN, Decision Model and Notation, Probabilistic Reasoning

Bio: Simon Vandevelde is a 4th year PhD researcher at KU Leuven (De Nayer campus) in Belgium, as part of the EAVISE/DTAI research group. His research is focused around user-friendly knowledge representation languages, as part of which he has been in investigating the intersection between DMN and Knowledge Representation and Reasoning. Research results so far include a constraint-based extension to DMN (cDMN), a probabilistic extension (pDMN), and a novel, “context-aware” approach to decision table verification.

A solution to the Roadef 2022 Challenge by Helmut Simonis (Insight Centre for Data Analytics, School of Computer Science and Information Technology, University College Cork)

The Roadef Challenge for the year 2022 was posed by Renault, and focusses on truck transport from suppliers to plants for their global supply chain. The task is to deliver all required materials with a minimum number of truck movements, while also reducing early delivery in order to minimize inventory cost. Items with the same footprint can be stacked on top of each other, respecting constraints on cardinality, height and weight, while stacks must be placed in the trailers so that each stack is supported to the front by other stacks, and the non-linear axle weight constraints of the truck/trailer combination are respected. Data for 120 instances are provided, each using up to 6000 trucks and carrying 4.5 million pieces over a period of several months. We present our solution to this problem, which is based on an overall problem decomposition, using a Mixed Integer Programming solver to assign items in time, while solving individual optimization models to build stacks and place them in each trailer. We find solutions for each of the given 120 instances in less than one hour.
Keywords: supply chain management, logistics, truck transport, placement problems, problem decomposition

Helmut Simonis is a senior research fellow at the Insight Centre for Data Analytics in Cork, Ireland, working on the combination of Machine Learning and Optimization. He has contributed to the area of Constraint Programming since 1986, has been involved in multiple start-up companies, and is a past president of the Association for Constraint Programming. Current research interests are automatically learning constraint models from example solutions, and improving decision making under uncertain forecasts.

Decision model, meet the real world: Testing optimization models for use in production environments by Ryan O’Neil and Nicole Misek (NextMv.io)

When optimization technology works well, it feels magical. But it is not magic. Good decision optimization is both art and science. But when models are released into the wild, they often behave differently than expected. How do we tame our automation and make it predictable?
Similar to MLOps, a diverse suite of testing methodologies such as acceptance testing, shadow testing, scenario testing, and switchback testing help us master change and operational risk. We explore common testing techniques in the context of an on-demand delivery marketplace, optimization models and the real world interact in interesting ways.
We cover real-world examples from on-demand delivery, including:
* Vehicle routing
* Driver scheduling
* Demand forecasting
We demonstrate some of these testing techniques on these optimization models, including historical tests and acceptance tests. We present these not as isolated tools, but as a cohesive toolbox to build confidence in optimization models and achieve better results in the real world.
Keywords: testing, optimization, routing, scheduling, mlops, orops

Ryan O’Neil led the Decision Engineering department at Grubhub and Zoomer, which owned forecasting, scheduling, routing, and simulation. Ryan worked as an Operations Research Analyst at MITRE, and led software teams at The Washington Post, Yhat, and Polimetrix. During this, he earned a PhD in Operations Research at George Mason University, and wrote his dissertation on real time routing for pickup and delivery problems.

Nicole is VP of Engineering at Nextmv. Previous to Nextmv, Nicole worked as a statistician at Grubhub where she helped the scheduling, dispatch, and market management teams improve their systems. Outside of work, Nicole spends most of her time chasing around her two toddlers and keeping them from getting into too much trouble

A practical guide to a deployment under uncertainty by Carole-Ann Berlioz (Sparkling Logic)

Although the bulk of the work resides in authoring decision logic, the work is not done once this talk is completed. Companies require a formal process to control the promotion of their strategies from development to testing, and ultimately to production. Technology makes it easy to turn business rules and predictive models to an executable service. Yet, there is more art than science to the task of deployment. While teams rely on extensive unit and regression testing, uncertainty is still looming. How can you improve your odds that a newly deployed strategy will perform? There are indeed techniques that can help you. In this presentation, Carole-Ann will present a spectrum of techniques that can be combined to instill greater confidence in the quality of candidate strategies. Starting in the development sandbox with simulations, moving on to canary deployments, and ending with actual or shadow champion / challenger experiments, Carole-Ann will cover the benefits of each technique, with practical tips on how to put them in place effectively. Keywords: Practitioner, Real-Life Experience, Deployment

Carole-Ann Berlioz is Co-Founder and Chief Product Officer at Sparkling Logic, a leading Decision Management Platform vendor known for its innovation. Over a few decades, she has led product management and strategy for generations of award-winning business rules, predictive analytics, and optimization products. In 2010, she teamed up with Chief Architect and CTO Carlos Serrano-Morales to create Sparkling Logic, a “Cool Vendor” that has gained momentum around the world, uniquely serving Business Analysts with an intuitive yet comprehensive, fully integrated decision manager, SMARTS™. In addition to her visionary role, she also takes pride in building client projects for financial services and insurance companies. Her hands-on expertise fuels her creativity in this industry, recognized with several patents in Decision Management and Adaptive Analytics.

An intelligent information system for pension communication – the Belgian case study by Myriam Lanotte, Manuel Kolp and Pierre Devolder (UCLouvain, Belgium)

Over the past 10 years, due to the increasing efforts in digital transformation and e-government, European countries have invested in the development of digital platforms-specific pensions. These e-platforms are crucial for the future of pension communication, information, and knowledge management. The development of these platforms has to face many challenges: complex systems, frequently changing environments, dealing with citizens’ expectations and needs, with stakeholders’ considerations and expectations, respecting the legislation, difficulties to reach young people, dealing with different pension phases, dealing with different citizens profiles, low pension knowledge, low financial literacy, people behavior with respect to pension. Many efforts such as dedicated workshops and task forces involving national authorities, OECD, the European Commission, and other European organizations have been organized recently around the world to deal with substantial questions tackling the strategic goal (“What is a good digital pension platform?”), tactical structure (“What are the needed functionalities?”), operational aspects (“how to design such platform?”), or behavioral processes (“how to deal with the above-listed challenges”?). In this article, we present the benefits of using a goal and agent-oriented requirements engineering methodology, the i* framework, to answer these questions and to model the functional and non-functional requirements of a digital pension platform able to deal with so many challenges. We use the Belgian pension case to illustrate the application of the methodology but the presented methodology is generic enough to be applied in any country and pension digital platform.
Keywords: Requirement analysis, Agents, Agent-Oriented analysis, I* framework, Digital pension platform, Pension communication

After her Master in Business engineering specialized in technologies and information management, Myriam Lanotte started to work as IT analyst on the governmental database of occupational pensions in Belgium. Since then, she has specialized in pensions and become IT Team Leader. In 2019, she decided to start a Ph.D. to mix her expertise in pensions and Information management. The objective of her research is to explore pension communication and more particularly digital pension communication and intelligent systems to optimize people’s pension knowledge and ability to plan their pension.

DSL-Based Approach on Business Rules Unit Testing by Rimantas Zukaitis (EIS, Lithuania)

One of the challenges to overcome in the adoption of the business rules approach in enterprise information systems is the testing of implemented business rules. This becomes a real pain point when a business rules engine is used to implement data validation in complex business domains, for example, insurance. Typically there are hundreds or even thousands of data elements in the more complex domain models, the amount of validation rules – and interdependencies of them quickly grows beyond what can be covered with manual or regression testing.
The two most common approaches – application endpoint testing and implementation of unit tests using common programming tools bring their own set of issues. Endpoint testing typically tests application behavior as a whole – and does not provide a sufficient level of isolation to test and introspect the behavior of a particular rule of interest. Implementing unit tests in code can provide the necessary level of isolation and means to prepare and manipulate test data – but it causes the anti-pattern where the code (unit tests) depends on the configuration (rule definitions).
These issues are even more exacerbated when rule versioning and variability by dimensions are taken into account – the amount of test data and unit test fixtures can grow exponentially.
In this presentation, we will examine these challenges in detail and will demonstrate how they can be addressed by introducing a rules testing framework, based on Domain Specific Language (DSL) to define unit tests, and test cases and manage complex sets of testing data using no code or low code approach.
Keywords: rule testing, data validation rules, test management, test data management

With a background in fundamental Computer Science and Software Engineering, Rimantas Zukaitis is a Lead Software Architect for EIS. For the last 15 years, he has been focusing on the use case of applying business rules approach for data validation as well as development processes, models and tools to facilitate this approach. Over his career he worked on several generations of rule engine implementations, and currently is leading the team working on Kraken Rule Engine – a lightweight business rule validation engine for distributed architectures, allowing the application of the same business rules both on the backend and frontend.

Leveraging Decision Microservices to optimize Consultative Sales Processes by Adithya Buddhavarapu, Sriharsha Sanagapallli and Venkat Yerramsetty (Focal CXM)

When sales teams are in front of customers, assessing them in real time, scoring them and presenting right content and assessments is a challenging task to marketing and sales professionals because each customer can potentially fall into different segments and at different phases of the journey. At GoBaton, we have come up with an innovative solution to leverage Decision Micro Services to address the aforementioned problem.

With our solution, sales teams can flip open an iPad, select a customer or a lead and the system automatically picks the right assessment or content to be presented to the end-customer. When there is a churn in sales teams or new members come onboard, leveraging Rules Engines and Decision Making Services can take the burden of thinking from sales teams and helping them capture useful insights from customers. The scoring models can help categorize opportunities based on the results and move them accordingly through the funnel. Keywords: CRM, Salesforce, Automated Decisions, Mobile Decisions, Dynamic, Lead Scoring, Opportunities, Customers

Adi Buddhavarapu is the co-founder and CEO of FocalCXM, a company passionate about delivering remarkable customer experiences. Prior to founding Focal, Adi worked at Oracle Corporation as a Product Strategist where he was part of a team that developed world-class enterprise software for the global Life Sciences industry. He has decades of experience in CRM/CX Systems, Enterprise Content, Cloud and Data. 

SriHarsha is Senior Mobile Developer with experience in serving Commercial & Clinical Teams across various industries with online and offline applications to boost their productivity in the field. His expertise is around iOS and ReactNative Platforms. 

 

Venkat Yerramsetty is a Technical Lead at FocalCXM. He has vast experience in Software Engineering and leveraging Rules Engines to solve complex CRM problems around Events, Rebates and other workflows that impact compliance and employee productivity.