Page 8 of 14

END-USER PERSONALIZATION

Importance to Manufacturing

The Cockpit's target users belong to such domains as manufacturing, planning, logistics, materials, finance, and human resources. These users play different roles within each of these domains. For example, the manufacturing area includes factory and site managers, process engineers, quality engineers, and others with each role viewing data from a different perspective. The OLAP solution offered by Cockpit is optimal for slicing and dicing data, allowing users to pinpoint views of interest. It provides factory-level output figures to factory managers and provides process engineers interested in the same output with detail-level information for key constraint equipment.

Personalization Engine

Instead of providing a fixed number of static reports categorized by user types/roles, we elected to enable dynamic report generation, usage and retrieval. To support this functionality, the system should remember each user and maintain a user-specified personal configuration store. The personalization engine, a middle-tier service based on Site Server, handles personalization in the Cockpit. All registered Cockpit users have a dedicated account defined in the Site Server database with each user account holding four data segments:

User Defined Views: Users can create their own reports on-the-fly and save them for later retrieval. The user's personalization store saves, in XML format, only the necessary connection and parameter information. When a user selects a saved view, XML is retrieved from the personalization store and creates the report using the most recent OLAP cube data.

Front Page Views: Users have the ability to specify which of their favorite views will be displayed on the Cockpit's front page. Generated on the server, these views are rendered as images and served to the user's machine on demand.

Subscription Views: Users can receive selected views as images by e-mail or desktop channels.

User-Defined Exceptions: The normal trend is for users to analyze various static or active views and determine the current state of operations. Manual in nature, this process consumes valuable time. Instead of requiring users to scan reports and make decisions, the Cockpit's user-defined exception facility is designed to automate the report-checking process. Users have the ability to set exceptions at the indicator level, instructing the system to check for exceptions each time the data set changes. The system notifies the user via e-mail or desktop channels when it detects an exception, eliminating the need for manually checking of reports to identify the exception.

Scanning and Notification

The Indicator Reporting module excels at providing easy access to a wide array of data in the OLAP database. However, all data provided by the tool exists within the company, is quantitative, and requires users to proactively search for information. The Scanning and Notification module was designed to compliment the Indicator Reporting module by scanning the business environment for relevant qualitative and quantitative information, and pushing this information to the user in real-time.

Figure 14: Scanning and Notification module architecture

Scanning Engine
The scanning engine is an automated agent that monitors a variety of sources for information on behalf of the user. The information gathered by the agent is passed to the user through a variety of channels. The two primary types of information monitored by the agent are Web-based news feeds and internal databases. News feeds are pre-defined by a system administrator and organized into public channels with sub-channel categories. If the channel is of interest, the user can subscribe to it.

For example, there is a channel that monitors news feeds for information regarding our largest customers. The agent delivers news headlines to the user for scanning. If an article is of interest, the user can click through to the complete text. Additionally, the tool can monitor internal databases, including OLAP cubes behind the Cockpit Indicator Reporting module, for a mixture of personalized and predefined trigger points.

For example, if the user is always interested in the yield of a single product at a manufacturing site, that user can create and track the exception. An exception wizard will walk the user through the process of choosing which conditions to monitor. Generally, an exception can be triggered if it crosses a fixed value or deviates from a goal by a certain percentage. Assuming all the necessary exceptions are being monitored, users do not need to poll the Cockpit's Indicator Reporting module to ensure the factory is running within acceptable limits. Instead, managers can be assured that they will be notified when they need to take action.

Delivery Channel
Once the information is found, it needs to be pushed to the user in an easily consumable, timely way. Since type of information varies by user, the Cockpit provides a variety of methods for pushing this information to the user. The first method, pushing information through e-mail, is appealing since users are comfortable with the technology, and it provides a fairly real-time delivery mechanism.

Figure 15: The Cockpit Scanning and Notification module, desktop event delivery mechanism

The downside is that important notices may get lost in the Inbox along with the huge volume of mail. Additionally, it is not very efficient for news headline scanning. To overcome these shortcomings, a second method is offered that provides a faster response time and information categorization, but requires a new application.

This application runs as a desktop event delivery mechanism organized according to channels and categories mentioned earlier. It uses a true push mechanism that notifies the user immediately when new information is available just by changing icon colors and alert lights on the application. Users can view this information at their convenience. These two choices are just the first of many that may become available.

Future implementations could easily include such channels as a Web portal or public folders. Since information agent and delivery channels are separate, this same information could be pushed through almost any channel. Users will be able to select the channel which is right for their choice of information and is the most comfortable.