Timing of anonymization rules in order to capture AFN digitization analytic reports

See Best Practices for Decision Pages and Tags for groups

Legend: not started IN PROGRESS STALLED decided

Status	STALLED
Description	Timing of anonymization rules in order to capture automated fulfillment network (AFN) digitization analytic reports
Decision summary	In order to gather needed reports on automated network fulfillment activity from Alma Analytics the “days since loan ended” in the Loan Anonymization Rules of the fulfillment anonymization job should be set to at least 120 days. RS OST will investigate other workaround (additional user statistical category) with Ex Libris for this defect in their Alma Analytics
Work Plan Deliverable	Resource Sharing Analytics Project Team, Phase 1 (Alma support for UCOP statistics schedule c)
Owning group	Resource Sharing Operations SubTeam + sils-rs-l [@] ucop.edu
Approver	Operations Team
Consulted	Fulfillment OST, Data Privacy and Security Project Team, Operations Team
Informed	Cohort
Decision-making process	RSOST drafts recommendation and solicits feedback, and presents final recommendation to OT for endorsement.
Priority
Target decision date	(pending feasibility investigation of another workaround with Ex Libris)
Date decided	[type // to add Date]

Recommendation

In order to gather needed reports on automated network fulfillment activity from Alma Analytics the “days since loan ended” in the Loan Anonymization Rules of the fulfillment anonymization job should be set to at least 120 days.

Reasoning

It is necessary to run AFN reports through Alma Analytics; however, the Analytics database structure is dependent on links to patron record in order to compile the reports that include digitization requests. Anonymization jobs in Alma break the link between the patron and the requests - and once that link is broken it’s no longer possible to effectively get accurate reports.

Due to the above flaw in the structure of digitization requests in the automated fulfillment network, we recommend that the “days since loan ended” in the Loan Anonymization Rules of the fulfillment anonymization job to 120 days so that needed reports can be gathered from Alma analytics. No reports contain personally identifiable information, however, the nature of the flaw removes all of the borrower portion of the digitization transaction data.

The likelihood of Ex Libris fixing the flaw in the database is low. So while this solution is a “workaround,” in practical terms this is likely to be necessary for the foreseeable future.

Impact

Stakeholder group	Impact

Stakeholder group	Impact
Resource Sharing OST	Loss of information on AFN requests negatively impacts needed reporting from Alma for interlibrary loan reporting needs, including (but not limited to) reports to support schedule c of the University of California Library Statistics data (aka “UCOP stats”).
Fulfillment OST	This change will impact anonymization on local requests in each IZ (as the AFN requests are the same sort of entity as local fulfillment requests).
Operations Team	The task of changing the appropriate anonymization rules would fall to this group.
Data Privacy and Security Project Team	There is a business need to have accurate automated fulfillment network reports, and therefore to keep patron information linked to borrowing requests as long as is necessary to run these reports. DPSPT policy is met as long as anonymization jobs are run regularly thereafter.
UCOP Administrative statistical needs	The compromised data encompasses part of the needed https://libraries.universityofcalifornia.edu/about/statistics-reporting-materials-for-the-university-of-california-libraries/ (schedule C) as gathered by University of California, Office of the President. Although the RS OST found a workaround to gather comparable numbers to support the 2021/2022 statistics, these reports would be more efficiently and easily gathered without this workaround.

Background

Anonymization jobs are performed at each IZ to remove any PII from transactions. These jobs are not currently harmonized - meaning, each campus has chosen to do these jobs on a different timeline, if at all. The RSOST is recommending for those IZs that run the fulfillment anonymization job to harmonize the anonymization job as above.

The Resource Sharing OST met with Ex Libris concerning analytic reports for AFN transactions in early August concerning various issues in gathering analytic reports for automated fulfillment network (AFN) transactions from Alma Analytics. During this meeting, Ex Libris stated that for loan transactions, the anonymization job breaks the link between the patron and the transaction. This removes all borrower information from the transaction and no borrower information is recoverable for that loan. Unfortunately, for loans taking place in the automated fulfillment network, rather than local loans, breaking the link on these transactions removes access to the needed non-PII information of the request.

For requests that do not generate a loan to the patron, such as requests that are filled digitally for the patron (such as article or book chapters), removing the patron link removes all information about the borrowing library on these requests. The anonymized AFN digitization requests are not retrievable within Alma Analytics.

Documentation

Ex Libris Documentation “Configuring Fulfillment Jobs” & its subsection “Configuring Anonymization”

(screenshots below from ExLibris Academy Anonymization session):

Location of ‘Fulfillment Jobs Configuration’

Location of ‘Loan Anonymization Rules’

Action Log

Action/Point Person	Expected Completion Date	Notes	Status

Action/Point Person	Expected Completion Date	Notes	Status
Consulting groups give feedback/endorsement	Sep 30, 2022		DONE
RS OST discuss & incorporate feedback into decision page	Oct 5, 2022	2022-10-05 RS OST meeting notes	done
RS OST final vote on decision			not started
RS OST presents to OT for endorsement & to make changes in IZes			not started