Now, there are actually professional consultants and probably a few organizations that can help guide you through an assessment like this one but you are likely to find examples like this available on the Internet. This one is a bit better than most but if I had to issue a letter grade to this assessment I would give it a C or a C+.
The first problem I have with this assessment is that it principally uses a reference that is over ten years old ("Paszko, C. & Pugsley, C., 2000"). When I was in school we would need to provide some special justification for using any resource older than ten years. That being said, the author took the time to actually list the references cited.
First off, chain of custody and unique sample identification are not the same thing so "provides a complete chain of custody" is a bit confusing. Unique sample identification does not create a chain of custody. When sample receiver gets the sample the LIMS should create an entry in the chain of custody mechanism, likely a record in some table consisting of at least the operator's unique id and the sample identifier plus the date/time/location of the sample. When the sample is transferred to a central place to be delivered to the lab that, again, will generate some kind of record in chain of custody with the same data. The cycle continues until disposition (or sending the sample back to its source if needed). There is no new sample identify created at each step -- it just gets created one time, typically when the sample is registered in the system. So, the fact that a sample has a unique identifier does not mean that any mechanism in the LIMS is tracking the location or use of a given sample. The identifier makes it possible to do this, but much more is needed.
On the same topic, some laboratories have no need for chain of custody so they either turn the feature off in their LIMS or simply ignore it. Chain of custody is not a central part of a LIMS, but rather, is peripheral functionality. As such I would de-emphasize its presence/operation during a LIMS assessment. Here is a short list of the primary LIMS functions:
- Single/Multiple sample registration (both automated via triggering/calendaring and manual)
- Unique sample identification
- Defined Groups/Roles for laboratory operations that control access to data/functionality
- Test Specifications
- Sample Receiving
- Labels w/Barcoding
- Template and Ad-Hoc sample definition and generation methods
- Sample routing feature
- Work Assignment feature
- Instrument interfacing
- Sample approval and data verification features
Depending on the industry these will make up the 'must have' list and some elements that would normally be considered secondary features like chain of custody, electronic data transfer, chemical inventory, and even QA/QC can move up to the primary tier. For instance, you can be certain that chain of custody is a required feature in a forensics LIMS.
Data entry is not limited to simply handling results. In a LIMS you want to capture customer information, instrument data, worksheet information, financial information, etc. Data entry is a valid thing to look at by itself however because when systems do a poor job of implementing it you will feel that pain every time you open the system. I've tested some LIMS and found that the data entry experience is painful, very painful, when the implementer forgot that the end user will need to repeatedly enter certain values on the same screen (consider sample receiving where an analyst, armed with only a barcode scanner, must log in hundreds of samples). Yes, you can implement data entry incorrectly in a LIMS. Also, reporting data back to clients falls out of scope for this issue.
Scheduling samples is a process different from automatically logging them in via a scheduler. They must be considered separately because nowadays humans and other software systems want to schedule samples in your LIMS. Two different components interact here (human/computer requester and some internal system that makes good on the request) to create an outcome. A good scheduler is one that is easy to understand and handles all of the types of scheduling you need. For instance, good ones handle holidays, days off, etc. automatically. Poor ones don't. There is no sense in lumping these together. The scheduler is complex enough, please don't lump it together with other utilities.
Some of the best LIMS simply provide an API to register samples and tests and a basic, stock scheduler. This way you can add/change and reconfigure the scheduler in a way that works for you, even if that means plugging it into a completely separate piece of software. With scheduling it can be a great idea to just get out of the user's way.
The rest of the article looks okay. I do not think this was meant to compete with a professional LIMS assessment. If it was I recommend that the consumer seek a second opinion.Go Back
Citation: Enviro-LIMS Assessment. (2011). Retrieved Mon May 1 02:06:14 2017, from http://www.limsexpert.com/cgi-bin/bixchange/bixchange.cgi?pom=limsexpert3;iid=readMore;go=1312843442