Difference between revisions of "User:Shawndouglas/sandbox/sublevel13"

From LIMSWiki
Jump to navigationJump to search
 
(49 intermediate revisions by the same user not shown)
Line 8: Line 8:
==Sandbox begins below==
==Sandbox begins below==
<div class="nonumtoc">__TOC__</div>
<div class="nonumtoc">__TOC__</div>
[[File:|right|400px]]
'''Title''': ''What are the key elements of a LIMS for animal feed testing?''


<!--{{LIMS Selection Guide for Manufacturing Quality Control/Taking the next step/Conduct initial research into a specification document tailored to your lab's needs}}//-->
'''Author for citation''': Shawn E. Douglas
==5. Taking the next step==
[[File:Specification leading to Design Documents.jpg|right|500px]]In section 3.4 of this guide, we briefly discussed how a user requirements specification (URS) fits into the process of purchasing [[laboratory informatics]] solutions for your manufacturing-focused [[laboratory]]. The URS has been viewed as a means for the purchaser to ensure their needs are satisfied by the functionality of the software. Traditionally, this has turned into a "wish list" for the purchaser, which while somewhat practical still lacks in its finesse. One common problem with this wishlist approach is the risk of "requirements creep," where more functionality than is truly necessary is desired, inevitably leading to a state where no vendor can meet all the wishlisted requirements. This makes selecting a solution even more difficult, particularly without significant prioritization skills.<ref name="AasemAnalysis10">{{cite journal |title=Analysis and optimization of software requirements prioritization techniques |author=Aasem, M.; Ramzan, M.; Jaffar, A. |journal=Proceedings from the 2010 International Conference on Information and Emerging Technologies |pages=1–6 |year=2010 |doi=10.1109/ICIET.2010.5625687}}</ref><ref name="Hirsch10Steps13">{{cite web |url=https://www.phase2technology.com/blog/successful-requirements-gathering |title=10 Steps To Successful Requirements Gathering |author=Hirsch, J. |publisher=Phase2 Technology, LLC |date=22 November 2013 |accessdate=07 December 2022}}</ref><ref name="BurrissSoftware07">{{cite web |url=http://sce2.umkc.edu/BIT/burrise/pl/requirements/ |archiveurl=https://web.archive.org/web/20190925003040/http://sce2.umkc.edu/BIT/burrise/pl/requirements/ |title=Requirements Specification |work=CS451R, University of Missouri–Kansas City |author=Burris, E. |publisher=University of Missouri–Kansas City |date=2007 |archivedate=25 September 2019 |accessdate=07 December 2022}}</ref>


Noting the potential problems with this wishlist approach, [[LII:LIMSpec 2022 R2|LIMSpec]]—a specification document for laboratory informatics solutions—took a new approach and turned to standards and regulations that drive laboratories of all types, as well as the data they manage. LIMSpec was rebuilt based on [[ASTM E1578|ASTM E1578-18]] ''Standard Guide for Laboratory Informatics'', as well as dozens of other standards and regulations, while still leaving room for a software buyer to add their own custom requirements for their industry or lab.  
'''License for content''': [https://creativecommons.org/licenses/by-sa/4.0/ Creative Commons Attribution-ShareAlike 4.0 International]


The rest of this chapter examines the research, documentation, and acquisition process that manufacturing labs needing laboratory informatics solutions will want to go through, with an emphasis on the utility of a sound requirements specification. While LIMSpec is offered as a solid starting point, you don't strictly need to use LIMSpec to conduct this process; the information in this chapter can largely be applied with or without LIMSpec itself.
'''Publication date''': May 2024


==Introduction==


===5.1 Conduct initial research into a specification document tailored to your lab's needs===
A specification is "a detailed precise presentation of something or of a plan or proposal for something."<ref name="MWSpec">{{cite web |url=https://www.merriam-webster.com/dictionary/specification |title=specification |work=Merriam-Webster |publisher=Merriam-Webster, Inc |accessdate=07 December 2022}}</ref> This concept of a specification as a presentation is critical to the laboratory seeking to find laboratory informatics software that fulfills their needs; they "present" their use case with the help of a requirements specification, and the vendor "presents" their ability (or inability) to comply through documentation and demonstration (more on that later). However, even the most seasoned of presenters at conferences and the like still require quality preparation before the presentation. This is where initial specification research comes into play for the lab.


Your lab's requirements specification document will eventually be a critical component for effectively selecting a laboratory informatics solution. There are numerous ways to approach the overall development of such a document. But why re-invent the wheel when others have already gone down that road? Sure, you could search for examples of such documents on the internet and customize them to your needs, or you and your team could brainstorm how a laboratory informatics solution should help your lab accomplish its goals. LIMSpec makes for one of the more thorough starting points to use, though you could also use other structured documents that have been developed by others. For the purposes of this guide, we'll look at LIMSpec.
This brief topical article will examine ...


The version of LIMSpec included in Appendix 1 of this guide is a slightly tweaked version of the original [[Book:LIMSpec 2022 R2|LIMSpec 2022]] document, omitting a few of the specialty laboratory functions that aren't applicable to manufacturing laboratories. You'll note that it's divided into five distinct sections, with numerous subsections in each:
'''Note''': Any citation leading to a software vendor's site is not to be considered a recommendation for that vendor. The citation should however still stand as a representational example of what vendors are implementing in their systems.


* Primary Laboratory Workflow
==Feed testing laboratory workflow, workload, and information management==
** 1. Sample and experiment registration
A feed testing lab can operate within a number of different production, research and development (R&D&#59; academic and industry), and public health contexts. They can<ref name="WardObtain24">{{cite web |url=https://animal.ifas.ufl.edu/media/animalifasufledu/dairy-website/ruminant-nutrition-symposium/archives/12.-WardRNS2024.pdf |format=PDF |author=Ward, R. |title=Obtaining value from a feed/forage lab engagement |work=Florida Ruminant Nutrition Symposium |date=27 February 2024 |accessdate=22 May 2024}}</ref>:
** 2. Sample management
** 3. Core laboratory testing and experiments
** 4. Results review and verification
** 5. Sample, experiment, and study approval and verification
** 6. Reporting
* Maintaining Laboratory Workflow and Operations
** 7. Document and records management
** 8. Resource management
** 9. Compliance management
** 10. Instrument and equipment management
** 11. Batch and lot management
** 12. Scheduled event management
** 13. Instrument data capture and control
** 14. Standard and reagent management
** 15. Inventory management
** 16. Investigation and quality management
* Specialty Laboratory Functions (minus non-relevant industries)
** 17. Production management
** 18. Statistical trending and control charts
** 19. Agriculture and food data management
** 24. Scientific data management
* Technology and Performance Improvements
** 26. Instrument data systems functions
** 27. Systems integration
** 28. Laboratory scheduling and capacity planning
** 29. Lean laboratory and continuous improvement
** 30. Artificial intelligence and smart systems
* Security and Integrity of Systems and Operations
** 31. Data integrity
** 32. Configuration management
** 33. System validation and commission
** 34. System administration
** 35. Cybersecurity
** 36. Information privacy


These sections and subsections should be able to address most any requirement you have for your system. Of course, if something isn't covered by LIMSpec, you can always add additional requirements.  
*act as a third-party consultant, interpreting analytical data;
*provide research and development support for new and revised formulations;
*provide analytical support for nutrition and contaminant determinations;
*provide development support for analytical methods;
*ensure quality to specifications, accreditor standards, and regulations;
*develop informative databases and data libraries for researchers;
*manage in-house and remote sample collection, labeling, and registration, including on farms; and
*report accurate and timely results to stakeholders, including those responsible for monitoring public health.


During the initial research towards your URS, you won't have to include every requirement for when you approach potential vendors. Most vendors appreciate a more inviting approach that doesn't overwhelm, at least initially. You will want to go with a limited yet practical set of requirements carefully chosen because they matter to you and your laboratory the most. In fact, you'll want to wait until after participating in several software demonstrations before even considering your URS to be complete. (More on that in 5.3.1.) This naturally leads us to a discussion about the RFI process.
This wide variety of roles further highlights the already obvious cross-disciplinary nature of analyzing animal feed ingredients and products, and interpreting the resulting data. The human [[Biology|biological]] sciences, [[Veterinary medicine|veterinary sciences]], [[environmental science]]s, [[chemistry]], [[microbiology]], [[radiochemistry]], [[botany]], [[epidemiology]], and more may be involved within a given animal feed analysis laboratory.<ref>{{Cite journal |last=Schnepf |first=Anne |last2=Hille |first2=Katja |last3=van Mark |first3=Gesine |last4=Winkelmann |first4=Tristan |last5=Remm |first5=Karen |last6=Kunze |first6=Katrin |last7=Velleuer |first7=Reinhard |last8=Kreienbrock |first8=Lothar |date=2024-02-06 |title=Basis for a One Health Approach—Inventory of Routine Data Collections on Zoonotic Diseases in Lower Saxony, Germany |url=https://www.mdpi.com/2813-0227/4/1/7 |journal=Zoonotic Diseases |language=en |volume=4 |issue=1 |pages=57–73 |doi=10.3390/zoonoticdis4010007 |issn=2813-0227}}</ref><ref name="PFPLSWHumanAnim18">{{cite web |url=https://www.aphl.org/programs/food_safety/APHL%20Documents/LBPM_Dec2018.pdf |format=PDF |title=Human and Animal Food Testing Laboratories Best Practices Manual |author=Partnership for Food Protection Laboratory Science Workgroup |date=December 2018 |accessdate=22 May 2024}}</ref><ref name=":0">{{Cite journal |last=Wood |first=Hannah |last2=O'Connor |first2=Annette |last3=Sargeant |first3=Jan |last4=Glanville |first4=Julie |date=2018-12 |title=Information retrieval for systematic reviews in food and feed topics: A narrative review |url=https://onlinelibrary.wiley.com/doi/10.1002/jrsm.1289 |journal=Research Synthesis Methods |language=en |volume=9 |issue=4 |pages=527–539 |doi=10.1002/jrsm.1289 |issn=1759-2879}}</ref> Given this significant cross-disciplinarity, it's arguably more challenging for software developers creating [[laboratory informatics]] solutions like a [[laboratory information management system]] (LIMS) that has the breadth to cover the production, R&D, and public health contexts of animal feed testing. In fact, an industry lab performing [[quality control]] (QC) work for a company will likely have zero interest in public health reporting functionality, and a LIMS that focuses on QC workflows may be more highly desirable.  


That said, this Q&A article will examine LIMS functionality that addresses the needs of all three contexts for animal feed analyses. Understand that the LIMS solution your feed lab may be looking for doesn't require some of the functionality addressed here, particularly in the specialty LIMS requirements section. But also understand the broader context of feed testing and how it highlights some of the challenges of finding a feed testing LIMS that is just right for your lab.


<!--{{LIMS Selection Guide for Manufacturing Quality Control/Taking the next step/Issue some of the specification as part of a request for information (RFI)}}//-->
==Base LIMS requirements for animal feed testing==
===5.2 Issue some of the specification as part of a request for information (RFI)===
Given the above ...
In some cases—particularly if your organization is of significant size—it may make sense to issue a formal RFI or request for proposal (RFP) and have laboratory informatics vendors approach your lab with how they can meet its needs. The RFI and RFP are traditional means towards soliciting bidding interest in an organization's project, containing the organization's specific requirements and vital questions that the bidder should be able to effectively answer. However, even if your organization chooses to skip the RFI or RFP process and do most of the investigative work of researching and approaching informatics vendors, turning to a key set of questions typically found in an RFI is extremely valuable towards your fact finding.


An RFI is an ideal means for learning more about a potential solution and how it can solve your problems, or for when you're not even sure how to solve your problem yet. However, the RFI should not be unduly long and tedious to complete for prospective vendors; it should be concise, direct, and honest. This means not only presenting a clear and humble vision of your own organization and its goals, but also asking just the right amount of questions to allow potential vendors to demonstrate their expertise and provide a clearer picture of who they are. Some take a technical approach to an RFI, using dense language and complicated spreadsheets for fact finding. However, as previously noted, you will want to limit the specified requirements in your RFI to those carefully chosen because they matter to you and your lab the most.<ref name="HolmesItsAMatch">{{cite web |url=https://allcloud.io/blog/its-a-match-how-to-run-a-good-rfi-rfp-or-rfq-and-find-the-right-partner/ |title=It's a Match: How to Run a Good RFI, RFP, or RFQ and Find the Right Partner |author=Holmes, T. |work=AllCloud Blog |accessdate=07 December 2022}}</ref>  
What follows is a list of system functionality important to most any feed testing laboratory, with a majority of that functionality found in many vendor software solutions.<ref name="WardObtain24" /><ref name="PFPLSWHumanAnim18" />


Remember, an RFI is not meant to answer all of your questions. The RFI is meant as a means to help narrow down your search to a few quality candidates while learning more about each other.<ref name="HolmesItsAMatch" /> Once the pool of potential software vendors is narrowed down, and you then participate in their demonstrations, you then can broadly add more requirements to the original collection of critical requirements from the RFI to ensure those providers meet all or most of your needs. That said, be cognizant that there may be no vendor that can meet each and every need of your lab. Your lab will have to make important decisions about which requirements are non-negotiable and which are more flexible. The vendors you engage with may be able to provide realistic advice in this regard, based upon your lab's requirements and their past experience with labs. As such, those vendors with real-world experience meeting the needs of manufacturing laboratories may have a strong leg up on other vendors.
'''Test, sample and result management'''


Again, Appendix 1 of this guide includes a comprehensive specifications document called LIMSpec, from which you can draw the requirements that are most critical to be addressed in an RFI. If you have zero experience developing an RFI, you may want to first seek out various example RFIs on the internet, as well as some basic advice articles on the topic. Some websites may provide templates to examine for further details. Broadly speaking, if you're conducting a full RFI or RFP, you're going to lead with the standard components of an RFI or RFP, including:
*Sample log-in and management, with support for unique IDs
*Sample batching
*[[Barcode]] and RFID support
*End-to-end sample and inventory tracking, through to reporting and disposition
*Pre-defined and configurable industry-specific test and method management, including for bacteria (i.e., microbiology), heavy metals (i.e., chemistry), radionuclides (i.e., radiochemistry), and other substances
*Pre-defined and configurable industry-specific workflows, including for production, R&D, and public health contexts
*Configurable screens and data fields
*Specification management
*Test, sampling, instrument, etc. scheduling and assignment
*Test requesting
*Data import and export
*Raw data management
*Robust query tools
*Analytical tools, including [[data visualization]], statistical analysis, and [[data mining]] tools
*Document and image management
*Version control
*Project and experiment management
*Method and protocol management
*Investigation management
*Facility and sampling site management
*Storage management and monitoring


* a table of contents;
'''Quality, security, and compliance'''
* an honest introduction and overview of your organization, its goals and problems, and the services sought to solve them;
* details on how the RFI or RFP evaluation process will be conducted;
* basis for award (if an RFP);
* the calendar schedule (including times) for related events;
* how to submit the document and any related questions about it, including response format; and
* your organization's background, business requirements, and current technical environment.


Being honest about your organization, its informatics requirements, and its current technical environment upfront in the RFI or RFP will also ensure that the time spent on the process is optimized for all involved parties. Before submitting any RFI, your lab will want to conduct thorough internal research ensuring everyone understands what the current technology and processes are, and how you all want to shape that with the introduction or updating of laboratory informatics systems. (If your lab has limited to no experience with adding automation and informatics elements to a laboratory, you may want to read through laboratory informatics veteran Joe Liscouski's [[LII:The Application of Informatics to Scientific Work: Laboratory Informatics for Newbies|''The Application of Informatics to Scientific Work: Laboratory Informatics for Newbies'']] for further insight.) You'll also want to answer critical questions such as "who will be responsible for maintaining the solution and its security?" and "how will our processes and procedures change with the introduction or updating of informatics systems?". These and other questions make up your business considerations, which should also address the:
*[[Quality assurance]] / [[quality control]] mechanisms
*Mechanisms for compliance with ISO 17025 and HACCP, including support for critical control point (CCP) specifications and limits
*Result, method, protocol, batch, and material validation, review, and release
*Data validation
*Trend and control charting for statistical analysis and measurement of uncertainty
*User qualification, performance, and training management
*[[Audit trail]]s and [[chain of custody]] support
*Configurable and granular role-based security
*Configurable system access and use (i.e., authentication requirements, account usage rules, account locking, etc.)
*[[Electronic signature]] support
*Data [[encryption]] and secure communication protocols
*Archiving and [[Data retention|retention]] of data and information
*Configurable data [[backup]]s
*Status updates and alerts
*Environmental monitoring support
*Incident and non-conformance notification, tracking, and management


* acquisition and long-term maintenance budget;
'''Operations management and reporting'''
* diversity of laboratory services offered now and into the future;
* level of in-house knowledge and experience with informatics systems and automation;
* level of in-house, executive buy-in of informatics adoption; and
* need for additional vendor pre-planning.


One other note: make it clear in any issued RFI that it's strictly a request for information and not a guarantee to issue a contract with any respondent.
*Configurable dashboards for monitoring, by product, process, facility, etc.
*Customizable rich-text reporting, with multiple supported output formats
*Custom and industry-specific reporting, including certificates of analysis (CoAs)
*Industry-compliant labeling
*Email integration and other communication support for internal and external stakeholders
*Instrument interfacing and data management, particularly for [[near-infrared spectroscopy]] (NIRS) instruments
*Third-party software interfacing (e.g., LES, scientific data management system [SDMS], other databases)
*Data import, export, and archiving
*Instrument and equipment management, including calibration and maintenance tracking
*Inventory and material management
*Supplier/vendor/customer management
*Flexible but secure client portal for pre-registering samples, printing labels, and viewing results
*Integrated (or online) system help


==Specialty LIMS requirements==


<!--{{LIMS Selection Guide for Manufacturing Quality Control/Taking the next step/Respond to or open dialogue with vendors}}//-->
*'''Mechanisms to make data and information more FAIR''': Like many other disciplines, modern academic and industrial research of feed ingredient selection, feed formulation, and feed production is plagued by interdisciplinary research data and information (i.e., objects) "in a broad range of [heterogeneous] information formats [that] involve inconsistent vocabulary and difficult‐to‐define concepts."<ref name=":0" /> This makes increasingly attractive data discovery options<ref name=":0" /> such as text mining, cluster searching, and [[artificial intelligence]] (AI) methods less effective, in turn hampering innovation, discovery, and improved health outcomes. As such, research labs of all sorts are increasingly turning to the FAIR principles, which encourage processes that make research objects more findable, accessible, interoperable, and reusable. A handful of software developers have become more attuned to this demand and have developed or modified their systems to produce research objects that are produced using [[metadata]]- and [[Semantics|semantic-driven]] technologies and frameworks.<ref name="DouglasWhyAre24">{{cite web |url=https://www.limswiki.org/index.php/LIMS_Q%26A:Why_are_the_FAIR_data_principles_increasingly_important_to_research_laboratories_and_their_software%3F |title=LIMS Q&A:Why are the FAIR data principles increasingly important to research laboratories and their software? |author=Douglas, S.E. |work=LIMSwiki |date=May 2024 |accessdate=22 May 2024}}</ref> Producing FAIR data is more important to the academic research and public health contexts of feed testing, but can still be useful to other industrial contexts, as having interoperable and reusable data in industry can lead to greater innovation and process improvement.<ref>{{Cite journal |last=van Vlijmen |first=Herman |last2=Mons |first2=Albert |last3=Waalkens |first3=Arne |last4=Franke |first4=Wouter |last5=Baak |first5=Arie |last6=Ruiter |first6=Gerbrand |last7=Kirkpatrick |first7=Christine |last8=da Silva Santos |first8=Luiz Olavo Bonino |last9=Meerman |first9=Bert |last10=Jellema |first10=Renger |last11=Arts |first11=Derk |date=2020-01 |title=The Need of Industry to Go FAIR |url=https://direct.mit.edu/dint/article/2/1-2/276-284/10011 |journal=Data Intelligence |language=en |volume=2 |issue=1-2 |pages=276–284 |doi=10.1162/dint_a_00050 |issn=2641-435X}}</ref> Of course, all animal feed testing labs can benefit when, for example, FAIR-driven, internationally accepted vocabulary and data descriptors for mycotoxin contamination data are used in research and laboratory software.<ref>{{Cite journal |last=Mesfin |first=Addisalem |last2=Lachat |first2=Carl |last3=Vidal |first3=Arnau |last4=Croubels |first4=Siska |last5=Haesaert |first5=Geert |last6=Ndemera |first6=Melody |last7=Okoth |first7=Sheila |last8=Belachew |first8=Tefera |last9=Boevre |first9=Marthe De |last10=De Saeger |first10=Sarah |last11=Matumba |first11=Limbikani |date=2022-02 |title=Essential descriptors for mycotoxin contamination data in food and feed |url=https://linkinghub.elsevier.com/retrieve/pii/S0963996921007833 |journal=Food Research International |language=en |volume=152 |pages=110883 |doi=10.1016/j.foodres.2021.110883}}</ref> This leads into...
===5.3 Respond to or open dialogue with vendors===
*'''Support for standardized and controlled vocabularies''': By extension, this gets into the matter of improved interoperability of feed testing results from different laboratories, particularly government labs in different jurisdictions responsible for monitoring contaminates in animal feed.<ref name="AAFCOSACStrat22">{{cite web |url=https://www.aafco.org/wp-content/uploads/2023/07/SAC_Strategic_Plan_2023-2025.pdf |format=PDF |title=Strategic Plan 2023-2025 |author=The Association of American Feed Control Officials, Strategic Affairs Committee |publisher=AAFCO |page=14 |date=16 November 2022 |accessdate=22 May 2024}}</ref> The Association of American Feed Control Officials (AAFCO) Strategic Affairs Committee (SAC) highlight this in their Strategic Plan for 2023–2025, stating that in order to "promote and integrate laboratory technology, methods, quality systems, and collaboration in support of animal food safety systems," the different LIMS used across various states demand an integrated IT environment where "comparable results from different labs" can effectively be made.<ref name="AAFCOSACStrat22" />
If you went the route of the RFI, you hopefully received more than a few well-crafted responses. Your RFI presumably included a small but critical set of requirements that needed to be addressed, and the vendors who responded dutifully addressed those critical requirements. Even if you didn't send out an RFI, you at least did your own research about some of the big players in the laboratory informatics space, and you may have even opened an initial dialogue with a few of them. If all has gone well, you're now at the point where you've narrowed down the pool of vendors but still have a basket of them to continue dialogue with. (If you're not comfortably at this point after an RFI or engagements with multiple vendors, you may need to either reconsider the effectiveness of your RFI or engagements or enlist help from a knowledgeable and experienced consultant to help steer you back on-course.)


As dialogue continues with vendors, you'll have several points to address:
==Conclusion==


1. What do I want their [[laboratory information management system]] (LIMS) to do for me?
2. How does their solution fit into our previously discussed budget?
Regarding question one, you've already laid some of the groundwork for that with the help of your handful of critical requirements (and the associated research that went into developing them). Outside of those critical requirements, a laboratory informatics solution should also provide clearly definable benefits to how you operate your manufacturing laboratory. These expected benefits should tie in with your overall business mission and goals. Using a LIMS as an example, here are a few of the benefits a well-developed LIMS can provide to practically any laboratory. Whenever you go through the discovery process with a vendor, you'll be asking how their system provides these and other benefits through its functionality. A quality LIMS can provide<ref name="McLelland98">{{cite web |url=http://www.rsc.org/pdf/andiv/tech.pdf |archiveurl=https://web.archive.org/web/20131004232754/http://www.rsc.org/pdf/andiv/tech.pdf |format=PDF |title=What is a LIMS - a laboratory toy, or a critical IT component? |author=McLelland, A. |publisher=Royal Society of Chemistry |page=1 |date=1998 |archivedate=04 October 2013 |accessdate=07 December 2022}}</ref><ref name="SciCompRisksBens">{{cite journal |title=Industry Insights: Examining the Risks, Benefits and Trade-offs of Today’s LIMS |journal=Scientific Computing |author=Joyce, J.R. |issue=January/February 2010 |pages=15–23 |year=2010}}</ref>:
* increased accuracy: the minimization or elimination of transcription and other errors;
* streamlined processes: ensuring each process step in a protocol/method is completed in the proper order, with all requirements met, updating sample statuses automatically;
* automation: integration with instruments, allowing for automatic uploading of samples and returning of results;
* regulatory and standards compliance: functionality that aids with compliance, including reporting results to state and local authorities;
* data security: role-based, configurable, secure access to data, processes, reporting, etc.;
* flexible reporting: reporting tools that allows for the design and generation of certificates of authority and other reports to lab- and regulation-based specs;
* instant data retrieval: query tools for finding data instantly according to any criteria (date range, test, product type, etc.); and
* configurability and cost-effectiveness: a user-configurable system (as opposed to hard-coded, requiring development for any modifications) that is flexible enough to adapt to rapid changes in test volume and type over time, without breaking the bank.
As for the second question, budgeting is always a tricky topic, both internally and when discussing it with vendors. We already mentioned in the previous section that addressing the acquisition and long-term maintenance budget of your solution(s) must be addressed as part of your lab's business considerations. (And we already mentioned some cost considerations in 3.1.6; this discussion will add a few more points.) The fact that laboratory informatics systems like the LIMS come in all kinds of price ranges makes it difficult to judge if a given system, as priced, is appropriate for your lab and its budget. There are some basic cost realities associated with LIMS acquisition<ref name="CSolsHowMuch17">{{cite web |url=https://www.slideshare.net/CSolsInc/how-much-does-a-lims-cost-licensing-and-beyond-pittcon-2017-tech-talk |title=How Much Does a LIMS Cost? Licensing and Beyond |author=Rosenberg, H.J. |work=SlideShare |date=28 March 2017 |accessdate=07 December 2022}}</ref><ref name="CSolsSaving18">{{cite web |url=https://www.csolsinc.com/blog/saving-costs-with-lims/ |title=Saving Costs with LIMS |publisher=CSols, Inc |date=25 October 2018 |accessdate=07 December 2022}}</ref>, which will help you understand where the vendor price comes from, and how it figures into your lab's budget (though some of these concepts may also apply to other informatics systems).
:1. Vendor pricing is generally based on how many will be using the system. This can be measured in concurrent users (how many will be using the system at any one time) or named users (the number of total users who will ever use the system, by name). Additionally, laboratory informatics vendors increasingly offer the option of a [[Cloud computing|cloud-hosted]] subscription, which of course has the advantage of not requiring your own IT department, and allowing labs to defray cost over time, with little or no actual license fee. Think about your usage strategy and choose the pricing format that makes the most sense for you.
:2. Most costs are related to the work involved with installing, configuring, and migrating data to the system. Try to choose a solution that has what you need out of the box, as much as possible. The more customized or unique options you ask for up-front, the more it tends to cost, as extra items are a function of the time it takes developers to add them.
:3. "User-configurable" beats "vendor-configurable" on cost-effectiveness. Some vendors offer a free or low-cost option, but don't be fooled. They are in business to make money, and they are counting on the fact that you'll need to pay them to make things work, add necessary functionality, and provide support and training. If you can find a vendor who offers a genuinely user-configurable system, and whose manuals and other support materials are clearly helpful and available so that you can adjust things the way you want, when you want, then that will go a long way toward budget efficiency and longevity.
:4. Additional interfaces and reporting requirements cost money. If necessary, consider phasing in any additional instrument and software interfaces over time, as revenue eases cash flow. You can go live with your system operations more quickly, entering results manually until you can afford to interface your instruments one-by-one. This goes for reports as well; a simple reporting module that meets regulatory requirements will do. You can make your reports and other exportable documents more attractive later.
Ideally, your budget has room for roughly $40- to $80,000 minimum (including setup, training, interfaces, etc.) for a quality, full-featured professional LIMS or LIS, with $300 to $900 per month (depending on number of users) for ongoing subscriptions. At around five concurrent users, the economics start to favor purchasing perpetual licenses rather than paying for a subscription. Purchased licenses will also entail ongoing annual or monthly costs as well (e.g., maintenance, support, warranty for updates etc.) Subscriptions (if available) are generally aimed at smaller labs. If you will be growing and scaling up, it may be a great way to get started, but make sure you have the option to switch to perpetual licenses later.
With much of this information in hand, you're likely ready to move on to finalizing the requirements specification and choosing a vendor, but not before you've sat through a few highly useful demonstrations.
====5.3.1 The value of demonstrations====
[[File:ForUM demo (2659615090).jpg|left|360px]]A demonstration of laboratory informatics solution is an integral part of making your final decisions. The demo offers a unique and valuable opportunity to see in-person how data and information is added, edited, deleted, tracked, and protected within the context of the application; you can ask about how a function works and see it right then and there. Equally, it is an excellent time to compare notes with the vendor, particularly in regard to the critical requirement that were addressed in your RFI (or through direct communication with the vendor). You can ask the vendor in real-time to answer questions about how a specific task is achieved, and the vendor can ask you about your lab's system and workflow requirements and how you best envision them being implemented in the system (e.g., does this interface seem intuitive?).
A demonstration is typically performed online, which is useful for a couple of reasons, COVID-19 notwithstanding. First, it means you can schedule and reschedule at your convenience, with little in the way of logistics to arrange. Second, the demonstration session is likely to be recorded (verify this), so everyone is clear on what was promised and what wasn't, how processes were shown to work, etc. Additionally, you can later review parts you may have missed, forgotten, or not quite understood, and you can share it with others, who then also get a look at the proposed system in action.
Be careful about falling for the temptation of presenting a full URS or other specification document to the vendor during the demonstration. You'll want to wait until after participating in several software demonstrations to consider presenting your full specification document to the vendor, and that's assuming that you've grown enamored with their solution. By waiting to finalize your lab's requirements specification until after the demos, a common error is avoided: too often labs think the first thing they must do is create a requirements list, then sit back and let the informatics vendors tell them how they meet it. Remember that even though most labs thoroughly understand their processes, they likely don't have as strong a grasp on the informatics portion of their processes and workflows. Participating in a demo before finalizing your list of specified requirements—or having only a minimal yet flexible requirements list during the demo—is a great way to later crosscheck the software features you have seen demonstrated to your lab's processes and any initial requirements specification you've made.<ref name="HammerHowTo19">{{cite web |url=https://www.striven.com/blog/erp-software-demo |title=How to Get the Most Value from an ERP Software Demo |author=Hammer, S. |work=The Takeoff |date=27 June 2019 |accessdate=07 December 2022}}</ref> After all, how can you effectively require specific manufacturing-related functions of your laboratory informatics software if you don't fully know what such an industry-specific system is capable of? After the demonstrations, you may end up adding several requirements to your final specifications document, which you later pass on to your potential vendors of choice for final confirmation.
<!--{{LIMS Selection Guide for Manufacturing Quality Control/Taking the next step/Finalize the requirements specification and choose a vendor}}//-->
===5.4 Finalize the requirements specification and choose a vendor===
Now that the demonstrations have been conducted and more questions asked, you should be close to finalizing your requirement specifications with one ore more vendors. In fact, you may have taken LIMSpec, chosen a few critical requirements from it, added them to a few unique requirements of your own, and included them as part of an RFI or question and answer session with vendors. You then likely took those responses and added them to your wider overall specification (e.g., LIMSpec), along with your own notes and observations from interacting with the vendor. This may have been repeated for several vendors and their offerings.
At this point, you're likely ready to either have those vendors complete the rest of the responses for their corresponding URS, or you may even be ready to narrow down your vendor selection. This all likely depends on what the initial fact finding revealed. How well did the vendors respond to your laboratory's unique set of needs? Were there critical areas that one vendor could address with their off-the-shelf solution but another vendor would have to address with custom coding? Did any of the vendors meet your budget expectations? Have you followed up on any references and customer experiences the vendors provided to you?
It may be that several vendors are appealing at this point, meaning it's time to have them respond to the rest of the URS. This makes not only for good due diligence, to better ensure most requirements can be met, but also a reviewable option for any "tie-breaker" you have between vendors. In reality, this tie-breaker scenario would rarely come up; more often, some other aspect of the software, company, or pricing will be a stronger limiter. However, you still want to get all those vendor responses, even if you've early on filtered your options down to one vendor.
Ultimately, your specification document may look similar to the LIMSpec, or it may have a slightly different format. Many prospective buyers will develop a requirement specification in Microsoft Excel, but that has a few minor disadvantages. Regardless of format, you'll want to give plenty of space for vendors to submit a response to each requirement. For your convenience, a Microsoft Word version of Appendix 1's LIMSpec for manufacturing labs is also included as part of this guide (see A8. LIMSpec in Microsoft Word format). That document is editable, giving end users and vendors the flexibility to remove information and enlarge columns.
Additionally, remember that often is the case that after the URS is completed and final questions asked, no single vendor can meet all your needs. Be ready for this possibility, whether it be a functionality requirement or a budget issue. Know ahead of time where your laboratory is willing to be flexible, and how much flex you have. After all of your lab's preparation, and with a little luck, you've found a vendor that fits the bill, even if a few minor compromises had to be made along the way.


==References==
==References==
{{Reflist|colwidth=30em}}
{{Reflist|colwidth=30em}}
<!---Place all category tags here-->

Latest revision as of 21:45, 22 May 2024

This is sublevel13 of my sandbox, where I play with features and test MediaWiki code. If you wish to leave a comment for me, please see my discussion page instead.

Sandbox begins below

[[File:|right|400px]] Title: What are the key elements of a LIMS for animal feed testing?

Author for citation: Shawn E. Douglas

License for content: Creative Commons Attribution-ShareAlike 4.0 International

Publication date: May 2024

Introduction

This brief topical article will examine ...

Note: Any citation leading to a software vendor's site is not to be considered a recommendation for that vendor. The citation should however still stand as a representational example of what vendors are implementing in their systems.

Feed testing laboratory workflow, workload, and information management

A feed testing lab can operate within a number of different production, research and development (R&D; academic and industry), and public health contexts. They can[1]:

  • act as a third-party consultant, interpreting analytical data;
  • provide research and development support for new and revised formulations;
  • provide analytical support for nutrition and contaminant determinations;
  • provide development support for analytical methods;
  • ensure quality to specifications, accreditor standards, and regulations;
  • develop informative databases and data libraries for researchers;
  • manage in-house and remote sample collection, labeling, and registration, including on farms; and
  • report accurate and timely results to stakeholders, including those responsible for monitoring public health.

This wide variety of roles further highlights the already obvious cross-disciplinary nature of analyzing animal feed ingredients and products, and interpreting the resulting data. The human biological sciences, veterinary sciences, environmental sciences, chemistry, microbiology, radiochemistry, botany, epidemiology, and more may be involved within a given animal feed analysis laboratory.[2][3][4] Given this significant cross-disciplinarity, it's arguably more challenging for software developers creating laboratory informatics solutions like a laboratory information management system (LIMS) that has the breadth to cover the production, R&D, and public health contexts of animal feed testing. In fact, an industry lab performing quality control (QC) work for a company will likely have zero interest in public health reporting functionality, and a LIMS that focuses on QC workflows may be more highly desirable.

That said, this Q&A article will examine LIMS functionality that addresses the needs of all three contexts for animal feed analyses. Understand that the LIMS solution your feed lab may be looking for doesn't require some of the functionality addressed here, particularly in the specialty LIMS requirements section. But also understand the broader context of feed testing and how it highlights some of the challenges of finding a feed testing LIMS that is just right for your lab.

Base LIMS requirements for animal feed testing

Given the above ...

What follows is a list of system functionality important to most any feed testing laboratory, with a majority of that functionality found in many vendor software solutions.[1][3]

Test, sample and result management

  • Sample log-in and management, with support for unique IDs
  • Sample batching
  • Barcode and RFID support
  • End-to-end sample and inventory tracking, through to reporting and disposition
  • Pre-defined and configurable industry-specific test and method management, including for bacteria (i.e., microbiology), heavy metals (i.e., chemistry), radionuclides (i.e., radiochemistry), and other substances
  • Pre-defined and configurable industry-specific workflows, including for production, R&D, and public health contexts
  • Configurable screens and data fields
  • Specification management
  • Test, sampling, instrument, etc. scheduling and assignment
  • Test requesting
  • Data import and export
  • Raw data management
  • Robust query tools
  • Analytical tools, including data visualization, statistical analysis, and data mining tools
  • Document and image management
  • Version control
  • Project and experiment management
  • Method and protocol management
  • Investigation management
  • Facility and sampling site management
  • Storage management and monitoring

Quality, security, and compliance

  • Quality assurance / quality control mechanisms
  • Mechanisms for compliance with ISO 17025 and HACCP, including support for critical control point (CCP) specifications and limits
  • Result, method, protocol, batch, and material validation, review, and release
  • Data validation
  • Trend and control charting for statistical analysis and measurement of uncertainty
  • User qualification, performance, and training management
  • Audit trails and chain of custody support
  • Configurable and granular role-based security
  • Configurable system access and use (i.e., authentication requirements, account usage rules, account locking, etc.)
  • Electronic signature support
  • Data encryption and secure communication protocols
  • Archiving and retention of data and information
  • Configurable data backups
  • Status updates and alerts
  • Environmental monitoring support
  • Incident and non-conformance notification, tracking, and management

Operations management and reporting

  • Configurable dashboards for monitoring, by product, process, facility, etc.
  • Customizable rich-text reporting, with multiple supported output formats
  • Custom and industry-specific reporting, including certificates of analysis (CoAs)
  • Industry-compliant labeling
  • Email integration and other communication support for internal and external stakeholders
  • Instrument interfacing and data management, particularly for near-infrared spectroscopy (NIRS) instruments
  • Third-party software interfacing (e.g., LES, scientific data management system [SDMS], other databases)
  • Data import, export, and archiving
  • Instrument and equipment management, including calibration and maintenance tracking
  • Inventory and material management
  • Supplier/vendor/customer management
  • Flexible but secure client portal for pre-registering samples, printing labels, and viewing results
  • Integrated (or online) system help

Specialty LIMS requirements

  • Mechanisms to make data and information more FAIR: Like many other disciplines, modern academic and industrial research of feed ingredient selection, feed formulation, and feed production is plagued by interdisciplinary research data and information (i.e., objects) "in a broad range of [heterogeneous] information formats [that] involve inconsistent vocabulary and difficult‐to‐define concepts."[4] This makes increasingly attractive data discovery options[4] such as text mining, cluster searching, and artificial intelligence (AI) methods less effective, in turn hampering innovation, discovery, and improved health outcomes. As such, research labs of all sorts are increasingly turning to the FAIR principles, which encourage processes that make research objects more findable, accessible, interoperable, and reusable. A handful of software developers have become more attuned to this demand and have developed or modified their systems to produce research objects that are produced using metadata- and semantic-driven technologies and frameworks.[5] Producing FAIR data is more important to the academic research and public health contexts of feed testing, but can still be useful to other industrial contexts, as having interoperable and reusable data in industry can lead to greater innovation and process improvement.[6] Of course, all animal feed testing labs can benefit when, for example, FAIR-driven, internationally accepted vocabulary and data descriptors for mycotoxin contamination data are used in research and laboratory software.[7] This leads into...
  • Support for standardized and controlled vocabularies: By extension, this gets into the matter of improved interoperability of feed testing results from different laboratories, particularly government labs in different jurisdictions responsible for monitoring contaminates in animal feed.[8] The Association of American Feed Control Officials (AAFCO) Strategic Affairs Committee (SAC) highlight this in their Strategic Plan for 2023–2025, stating that in order to "promote and integrate laboratory technology, methods, quality systems, and collaboration in support of animal food safety systems," the different LIMS used across various states demand an integrated IT environment where "comparable results from different labs" can effectively be made.[8]

Conclusion

References

  1. ↑ 1.0 1.1 Ward, R. (27 February 2024). "Obtaining value from a feed/forage lab engagement" (PDF). Florida Ruminant Nutrition Symposium. https://animal.ifas.ufl.edu/media/animalifasufledu/dairy-website/ruminant-nutrition-symposium/archives/12.-WardRNS2024.pdf. Retrieved 22 May 2024. 
  2. ↑ Schnepf, Anne; Hille, Katja; van Mark, Gesine; Winkelmann, Tristan; Remm, Karen; Kunze, Katrin; Velleuer, Reinhard; Kreienbrock, Lothar (6 February 2024). "Basis for a One Health Approach—Inventory of Routine Data Collections on Zoonotic Diseases in Lower Saxony, Germany" (in en). Zoonotic Diseases 4 (1): 57–73. doi:10.3390/zoonoticdis4010007. ISSN 2813-0227. https://www.mdpi.com/2813-0227/4/1/7. 
  3. ↑ 3.0 3.1 Partnership for Food Protection Laboratory Science Workgroup (December 2018). "Human and Animal Food Testing Laboratories Best Practices Manual" (PDF). https://www.aphl.org/programs/food_safety/APHL%20Documents/LBPM_Dec2018.pdf. Retrieved 22 May 2024. 
  4. ↑ 4.0 4.1 4.2 Wood, Hannah; O'Connor, Annette; Sargeant, Jan; Glanville, Julie (1 December 2018). "Information retrieval for systematic reviews in food and feed topics: A narrative review" (in en). Research Synthesis Methods 9 (4): 527–539. doi:10.1002/jrsm.1289. ISSN 1759-2879. https://onlinelibrary.wiley.com/doi/10.1002/jrsm.1289. 
  5. ↑ Douglas, S.E. (May 2024). "LIMS Q&A:Why are the FAIR data principles increasingly important to research laboratories and their software?". LIMSwiki. https://www.limswiki.org/index.php/LIMS_Q%26A:Why_are_the_FAIR_data_principles_increasingly_important_to_research_laboratories_and_their_software%3F. Retrieved 22 May 2024. 
  6. ↑ van Vlijmen, Herman; Mons, Albert; Waalkens, Arne; Franke, Wouter; Baak, Arie; Ruiter, Gerbrand; Kirkpatrick, Christine; da Silva Santos, Luiz Olavo Bonino et al. (1 January 2020). "The Need of Industry to Go FAIR" (in en). Data Intelligence 2 (1-2): 276–284. doi:10.1162/dint_a_00050. ISSN 2641-435X. https://direct.mit.edu/dint/article/2/1-2/276-284/10011. 
  7. ↑ Mesfin, Addisalem; Lachat, Carl; Vidal, Arnau; Croubels, Siska; Haesaert, Geert; Ndemera, Melody; Okoth, Sheila; Belachew, Tefera et al. (1 February 2022). "Essential descriptors for mycotoxin contamination data in food and feed" (in en). Food Research International 152: 110883. doi:10.1016/j.foodres.2021.110883. https://linkinghub.elsevier.com/retrieve/pii/S0963996921007833. 
  8. ↑ 8.0 8.1 The Association of American Feed Control Officials, Strategic Affairs Committee (16 November 2022). "Strategic Plan 2023-2025" (PDF). AAFCO. p. 14. https://www.aafco.org/wp-content/uploads/2023/07/SAC_Strategic_Plan_2023-2025.pdf. Retrieved 22 May 2024. 
Retrieved from "https://www.limswiki.org/index.php?title=User:Shawndouglas/sandbox/sublevel13&oldid=63688"