Decision Support for Request Processing

Follow
These release notes describe functionality that may not have been released yet. To see when this functionality is planned to release, please review What's new and planned for ILLiad. Delivery timelines and projected functionality may change or may not ship.

Overview

With Decision Support, ILLiad can be configured to do more than ever for you, starting with borrowing request processing. New decision pipelines are being added that will allow you to automate the borrowing requesting process for both articles and loans. Addons can run at different stages of the processes, allowing for data in the request to be updated, availability in different systems to be checked, and finally either automatically sending the request based on your own customizable business rules or leaving it for staff review - with all the information automatically gathered from different systems and making it as easy as a single click to send the request to any of the available systems!

To enable the new automator pipelines, you will simply enable a new routing rule that has been added to your Routing table. Requests that get to the new Awaiting Request Pipeline queue (or the Awaiting Pre-copyright Pipeline for Articles) will get automatically picked up and processed by ILLiad.

Three different automation stages are run to enable you to fully automate your requesting in many cases. Stage 1 will let you verify and add additional information to the supplied request. Stage 2 will let you check availability from different systems via addons. And in Stage 3, ILLiad will take the newly acquired information along with the request itself and evaluate it against business rules you've set up to determine where you want to send the item. 

Workflow Flowchart

FlowchartPipelineProcess.png

Features

Three Stages to each Pipeline Decision Point

  • Stage 1: Information gathering addons can run and add additional information to your request. For example, an addon that looks up additional information for a request based on a DOI number. These addons can run in parallel and will be able to wait on a prior addon to finish if desired before executing.
  • Stage 2: System availability, cost, and turnaround time gathering. These addons will execute in parallel using the current information in the request to determine availability. This information is stored in a new data table in ILLiad that each addon can write their information to called FulfillmentData.
  • Stage 3: Automator. The built-in automator runs a set of business rules which you define that ranks your options based on cost, availability, turnaround time, and a match string against the request to determine where it should be sent. Then, based on your setup, it will either send the request for you or place it into Awaiting Pipeline Request Review, with a new display area that shows all of your options and which one it recommends.

Separate Pipelines for Loans and Articles

  • Loans: Since loans don’t require a copyright review, there is a single decision point for the pipeline. The three stages run once, in order.
  • Articles: Since articles require a copyright step in some cases, there are 2 decision points: the 3 stages will run once pre-copyright and once post-copyright. Sites who chose to process copyright separately can move requests to the post-copyright pipeline decision point directly.

Use Routing Rules to send selected requests into the pipeline

Routing rules will not be automatically added to the Customization Manager. When the feature has been released, there will be an instruction guide with the minimum recommended routing rules to support the Decision Support pipeline.
  • Any request that you put into the Awaiting Request Pipeline queue (or Awaiting Pre-copyright Pipeline) will be handled by the business rules you’ve set up.
  • Manually routing requests to the Awaiting Request Pipeline queue will also work.

Pipelines defined by new Queues

  • Awaiting Request Pipeline - This queue is used for loans as well as articles that have passed the copyright step.
  • Awaiting Pre-copyright Pipeline - This queue is used for articles that require a copyright clearance step.
  • Awaiting Copyright Clearance - Pipeline - This queue is used for articles that have passed through the Pre-copyright pipeline and still require copyright review. This queue opens automatically on the Copyright Clearance Form.
  • Awaiting Pipeline Request Review - This queue is where any request that doesn’t get automatically sent by the automator goes. This can either be because none of the business rules matched or because the business rule it matched was set to Recommend vs. Automatic.
  • Awaiting Pipeline Sending - This queue is where requests are placed for the system to automatically send them based on your business rules.
  • Awaiting Pipeline Error Review - This queue is where requests are placed when an addon returns an error during the processing.
  • Pre-Copyright Pipeline Errors - This queue is where requests that are in the pre-copyright pipeline are placed when an addon returns an error. This queue is able to be opened into the Copyright Clearance Form.

New and Updated Forms for Processing Pipeline Requests

  • Awaiting Pipeline Request Review Form - this is a new batch-oriented form allowing you to see all the recommendations from the pipeline and easily send the requests to your different options.
  • Updated Copyright Clearance Form - this form has been updated to add the data returned from the Pre-copyright Pipeline addons in the FulfillmentData table.
  • Updated Request Form - the request form has been updated to add the data from the pipeline addons shows in the FulfillmentData table.

FulfillmentData table to store newly available request information

  • This table is written to by the addons running in stage 2 of the pipeline. Information is stored here to help with decisions based on business rules in Stage 3.
    • Source - this will usually be a system name like Rapid, Reprints Desk, etc.
    • Cost - the cost of requesting from this source for this item, in US dollars.
    • Turnaround Time - turnaround time for this item is available, in hours.
    • Format - a field for the item format that would be delivered.
    • Copyright Included - for articles, if copyright is included or not required for this item.
    • Custom Data - a field that the addon can write information that can be acted on in the business rules, consisting of a set of key/value pairs in a JSON format.
    • Transaction Number (TN)- The ILLiad reference number that allows a user to identify and locate a request.
    • Availability - a yes or no response as to whether the item can be supplied by the source. The information is stored to help track an addons success or failure for statistical data.
    • Notes - any special notes the addon received from the source provider. For example, taxes not included in pricing.

Business Rules

  • A new Customization Manager configuration page will allow setting of your automator business rules using a simple builder interface.
  • There are separate sets of rules for each possible pipeline: loans, pre-copyright articles, and post-copyright articles.
  • The rules are in a table, and the rules are processed in order from the first to the last. The first rule that matches your criteria is executed and the rest of the rules are ignored for that request.
  • The different fields of each row in the table are all added together and evaluated against the information gathered in Stage 2 and stored in FulfillmentData.
  • Leaving a field blank ignores that criteria when evaluating rules.
  • The rules consist of the following fields:
    • Pipeline - Current available pipelines are BorrowingArticlePreCopyright, BorrowingArticle, and BorrowingLoan.
    • RuleNumber - Order of rules within a pipeline.
    • RuleType - Automatic, Recommend, or Off.
      • Automatic - if this rule matches, it will automatically send the request.
      • Recommend - if this rule matches, it will note this in the system and stop the automator processing and send the request to Awaiting Pipeline Request Review.
      • Off - this rule will be ignored.
    • Target - this is a key/value pair that can include addons, system values, and routing:
      • System|Cheapest - the system will look at all available sources and return the cheapest one.
      • System|Fastest - the system will look at all available sources and return the fastest one.
      • Addon|AddonName - the system will use the referenced addon to process the request
      • Route|QueueToRouteTo - the system will route the request to whatever queue is referenced as QueueToRouteTo.
    • Match String - a match string that works like a routing rule match string, so that you can make more complex decisions based on both the information in the request and the user record, as well as the information in this table.
    • Notes - a description of the rule to help explain its purpose.

Automator Examples

Pipeline RuleNumber RuleType Target Match String
BorrowingLoan 1 Automatic Addon|Rapid u.Status = 'Faculty' and FulfillmentData.Cost <= 10
BorrowingArticlePreCopyright 1 Automatic Addon|ReprintsDesk u.Status = 'Graduate' and CustomData.Cost <= 20 and FulfillmentData.TurnaroundTime <= 8
BorrowingArticle 1 Recommend System|Cheapest FulfillmentData.Cost < 20 and FulfillmentData.TurnaroundTime < 6
BorrowingArticle 2 Automatic Addon|Rapid FulfillmentData.Cost < 20
  • BorrowingLoan - 1: This rule would match if you wanted to choose Rapid for any loan request where the cost was $10 or less and that the user was a faculty member and would automatically send the request.
  • BorrowingArticlePreCopyright - 1: This rule would send any request to Reprints Desk automatically if it was less than or equal to $20 and it was for an article for a graduate student that could be delivered in 8 hours or less.
  • BorrowingArticle - 1: This rule would match if you wanted the cheapest option for an article that could be delivered in 6 hours but didn’t cost more than $20, then put the request into Awaiting Pipeline Request Review for staff to make the final decision.
    • Notes: Since BorrowingArticle is a post-copyright pipeline, the system would add the copyright cost (if entered) to any source that doesn’t say that Copyright Included is true. So it evaluates each system’s real cost based on both values.
  • BorrowingArticle - 2: This rule would match if you wanted an article that was available in Rapid for $10 or less to automatically send the request to Rapid.

 Related Links

Questions?

If this article didn’t resolve your issue, please take a moment and answer a few questions to help improve our documentation:

Feedback