Exploring Predictive Intelligence and related frameworks available within the ServiceNow platform

Introduction

Organizations are constantly seeking ways to optimize their operations and deliver exceptional experiences to customers. Artificial Intelligence (AI) has emerged as a powerful tool, enabling businesses to unlock valuable insights, make data-driven decisions, and revolutionize their processes. In this blog post, we will explore some of the AI features available within the ServiceNow platform, focusing on predictive intelligence, in particular the classification and similarity frameworks by presenting some real-life use cases.

What is actually AI? AI refers to the simulation of human intelligence in machines, enabling them to perform tasks that typically require human cognitive abilities. AI has been with us for quite long time, but since the ChatGPT release, AI has never been more accessible to everyone of us. Through advanced algorithms and data analysis, AI systems can learn, reason, and make predictions or decisions. In the context of ServiceNow, AI enhances its capabilities by leveraging predictive intelligence frameworks, empowering businesses to proactively address challenges, optimize workflows, and deliver services better, more efficiently and faster.

What are the outcomes you can expect?

1. Reduction of Manual Work: Predictive intelligence automates the analysis of data and task routing, reducing the need for manual intervention and allowing teams to focus on higher-value tasks and reducing the reassignment rate. (Classification framework)
2. Automated Prioritization: By analyzing the data and associated attributes, predictive intelligence can automatically prioritize the work based on predefined criteria, ensuring that critical issues receive immediate attention. (Classification framework)
3. Identification of Knowledge Gaps: Predictive intelligence can identify knowledge gaps by analyzing patterns and identifying areas where additional documentation or training may be needed. This enables organizations to improve knowledge management and enhance overall service quality. (Clustering framework)
4. Accelerated Resolution: With AI-based recommendations, predictive intelligence can provide real-time suggestions to support the issue resolution. This accelerates the resolution process by providing relevant information and guidance to support teams. (Similarity framework)
5. Faster Detection of Major Outages: Predictive intelligence can detect major outages faster by analyzing patterns, trends, and correlations within the data model. This allows organizations to identify and address large-scale service disruptions promptly, minimizing the impact on customers. (Similarity framework)

Classification Framework - automated categorization, routing and prioritization of work

The classification framework in ServiceNow's predictive intelligence leverages machine learning algorithms. These algorithms analyze the characteristics and attributes of incoming data to assign them to specific categories or classes. This automated categorization helps streamline the whole process and ensures that the right teams are engaged promptly reducing the manual effort and human intervention.

Here's how the classification framework works within ServiceNow's predictive intelligence:

1. Data Preprocessing: Before applying classification algorithms, data preprocessing is performed to clean, transform, and prepare the data. This step involves handling missing values, normalizing data, and encoding categorical variables to ensure consistent and reliable results.
2. Model Training: Using historical data and associated attributes, the classification framework trains a machine learning model. The model learns from the labeled data, identifying patterns and relationships between attributes and their assigned categories and teams
3. Feature Selection: The classification framework identifies the most relevant features or attributes that contribute significantly to the classification process. This step helps improve the model's accuracy and efficiency by focusing on the most informative data points.
4. Model Evaluation: To ensure the effectiveness of the classification model, it is evaluated using evaluation metrics such as precision and coverage. This step helps determine the model's performance and fine-tune it if necessary.
5. Task Categorization and Assignment: Once the classification model is trained and validated, it is applied to incoming records within ServiceNow. The model automatically assigns the task categories based on their attributes, such as short description, description, category or service. This automated classification streamlines the routing process, ensuring that ServiceNow tasks such as Incidents, HR Cases, Customer Cases are assigned to the most appropriate support team for efficient resolution faster.

Real-Life Use Case: Incident Categorization and Assignment

1. Incident Creation: The user initiates the incident creation process by submitting an incident with a short description, describing the issue as "VPN is not working. Cannot connect to company network"
2. Trained Model Application: The classification framework leverages a trained model, which has learned from historical incident data, to automatically analyze the incident attributes and predict its category. In this case, the model predicts that the category for this incident is "network" based on the provided description.
3. Automated Assignment: Alongside predicting the category, the trained model also predicts the appropriate assignment group for the incident. In this example, the model assigns the incident to the "Global Network Support" assignment group.
4. Incident Routing: Since the incident has been correctly classified and assigned by the trained model, it can be automatically routed to the correct assignment group without any human intervention. The incident is now in the hands of the Global Network Support team, ensuring that it reaches the right experts who specialize in network-related issues.

Similarity Framework - help the teams to resolve the issues faster by linking them to similar issues from the past

Similarity frameworks enable ServiceNow to identify patterns and similarities between records, facilitating the identification of potential recurring issues or common root causes. By grouping similar records together, teams can gain insights into broader trends and take proactive measures to prevent similar incidents from occurring in the future. This feature is only available within the Operations Workspaces.

The similarity framework in ServiceNow's Service Operations Workspace offers the following capabilities:

1. Assignment Group Recommendations: By analyzing the characteristics and attributes of incoming incidents, the similarity framework can identify similarities with previously resolved incidents and recommend the most suitable assignment group for efficient resolution. This helps ensure that incidents are routed to the appropriate teams with the necessary expertise, streamlining the resolution process.
2. Service Impacted Identification: The similarity framework analyzes the data to identify patterns and relationships, enabling it to determine the specific services impacted. This information helps teams prioritize their response efforts, allocate resources effectively, and minimize the impact on critical business services.
3. Major Outage Detection: The similarity framework can detect major outages by recognizing patterns that indicate a significant impact on multiple users or services. This allows organizations to promptly escalate and address these issues with a heightened sense of urgency, ensuring swift resolution and effective communication to stakeholders.
4. Similar Open Incidents / Cases: The similarity framework identifies other open cases that share similar characteristics and attributes with the current case. By providing visibility into these similar records, teams can identify potential trends or patterns, collaborate on resolution strategies, and leverage shared knowledge to expedite incident resolution.
5. Resolved issues for Resolution Assistance: The similarity framework analyzes historical case data to identify resolved cases that bear similarities to the current case. By leveraging this knowledge, support teams can access past resolutions, workarounds, or best practices, accelerating the resolution process and improving overall service quality.

Real-Life Use Case: Speed-up the resolution of a service interruption with the similarity framework

1. Incident Creation: A user opens an incident and reports an issue with opening Microsoft Excel files. The attempt results in an error, indicating a problem with the file opening process.
2. Trained Model Application: The ServiceNow similarity framework, powered by a trained model, quickly analyzes the incident attributes, including the description and any relevant details. It identifies similarities with past incidents that have been successfully resolved.
3. Recommendation in Service Operations Workspace: The similarity framework presents a recommendation within the Service Operations Workspace, indicating that there have been similar incidents in the past. This recommendation draws the agent's attention to the incident's knowledge base or resolution information.
4. Accessing Past Incident Resolution: The agent, prompted by the similarity recommendation, accesses the incident resolved in the past that shares similarities with the current incident. This allows the agent to quickly review the previous incident's details, resolution steps, and any relevant information.
5. Identifying the Solution: By examining the past incident's resolution, the agent discovers that the solution to the problem is to clean the registry files and restart the PC. The registry data corruption is identified as the cause of the Excel file opening issue. As a result incident is resolved much faster.

AI, with its predictive intelligence capabilities and associated frameworks, is transforming the way businesses operate and make decisions. The classification and similarity frameworks, which we have explored today, together with the additional capabilities of regression, and clustering frameworks empower organizations to extract valuable insights, optimize processes, and enhance customer experiences. By embracing AI and its frameworks, businesses can gain a competitive edge, make data-driven decisions, and unlock the untapped potential within their data assets.

As the world becomes increasingly data-centric, it is crucial for professionals and organizations to embrace AI and leverage the power of predictive intelligence. By harnessing these capabilities, businesses can uncover hidden patterns, make accurate predictions, and drive innovation in their respective industries.