One of the first questions we are often asked when we first meet a potential new client is “How fast can you turnaround the samples?” This is not surprising considering the criticality of some of the equipment we monitor such as the gearboxes on mile long conveyors in mines with downtime costs being greater than $1 Million for a 2-hour shut down. Therefore, with maintenance decisions having to be so fast this then puts a demand for the oil analysis to be even faster.
So how do we measure the time oil analysis takes? There are two commonly used methods:
- Lab Turnaround – I.e. the time from the sample arriving at the lab to a report being created and either emailed or uploaded online. Laboratories tend to focus most of their efforts on keeping this well managed and is the most common assessment method a lab uses to assess their own performance. However, this is an artificial construct as even if the laboratory turns the sample around instantly the delays before and afterwards are still perceived as a laboratory delay to the end client.
- Total Time – This is the time from the sampler taking the sample to receiving the report. This includes many factors:
- Sampler Time – Time from when the Sampler takes sample to attempting to post it. This may be direct handing to courier, or taking to a mailroom or head office to post onwards.
- Time waiting for collection – This may be the time from the sampler dropping off in the mail room, or the time between sampling and e.g. the office booking and batching collections.
- Time with the courier – This is the time the courier physically takes to get the samples from one location to another.
- Lab Turnaround – See above.
- Post processing – This is the time the report sits in the client’s inbox, or remains unread on LearnOilAnalysis. If the reports all go into a central office or person, then the time for them to get around to that sample can also have an impact on perceived turnaround.
Although Labs focus on ‘their’ turnaround, end clients’ perception of turnaround includes all the processes and so delays in those areas may be perceived as a lab turnaround issue. In 2015 and 2016 LearnOilAnalysis went through their customer service records for any samples that the client either complained were taking too long, called more than once to check the status of a sample, or simply mentioned on the phone that they believed the report would be ready by now and in 98% all these cases the delays were related to issues outside of the lab turnaround such as samplers leaving bags of samples in their van for a couple weeks, clients using the cheapest mailing option possible to send samples leading to delays in samples arriving. Some case studies follow explaining some examples of these 98%, however, there is still 2% that did take place within the laboratory. The breakdown of the 2% is shown in the pie chart below.
Although this list only accounts for 2% of delays as 98% are not lab related it has some interesting findings.
Incorrect sampling information accounts for nearly half of delays. This can be down to stating the grade is one product when it is not and hence the sample gets retested to confirm the data before suggesting the client may have incorrectly completed the form.
Poor Sampling is the contamination in the sample at sampling makes the data not trends such as a post filtration sampling being dirtier than the pre-filtration sample. Waiting on clients to respond to clarify questions on suites or the sample is also a big cause of delays. Retesting is how we go above and beyond for our clients so that when we are flagging a sample as abnormal we confirm the data first, not always just by the agreed method list, but we confirm by multiple different methods and even add some extra tests at no additional cost, especially if it is close to a borderline or the diagnostician believes will help them give a better diagnosis, giving our clients absolute confidence to act on our data. Unsuitable samples – A client requiring a viscosity at 100’C whilst full of water normally would be cancelled (as the water would boil affecting the result, however, the lab may attempt to run by drying off the sample first adding extra time. Likewise, elemental analysis is not the same across all types of oils owing to something called the matrix effect. This can mean that certain synthetic lubricants either do not easily dissolve in the most common and legal to use lab solvents or put out the flame on the ICP. Sometimes some R and D to get one sample to run can take a week, meaning the sample is late simply because the sample is not suitable for the tests requested. Finally the one that customers assume is the cause of all delays is Capacity issues, which accounts for just 1% of the all the reasons that make up the 2% of lab related delays (i.e. 0.02% of all delays). This is so low as the lab have multiple spares on most tests meaning if one instrument fails there is always another, or we can divert staff onto manually performing some of the robotic prepping of instruments. True there are some tests that are more suspecible to capacity issues such as RPVOT in which the sample can be running for days and even with half a dozen instruments if a client decides to batch up a years worth of samples there will be impact on that particular test temporarily. Additionally, there are some natural disasters such as the factory that produced all the probes for one instrument type was washed away by a Tsunami (they are back up and running now and we have purchased more instruments from them recently) meaning when an instrument failed a week later the labs reservce capacity for that instrument was reduced.
|Case Study Turnaround (1) – Streamline your sampling process:
In Late 2016 an industrial food manufacturing plant client complained to the laboratory that the last 3 batches of oil sample turnarounds were unacceptable and that they were looking to switch suppliers if things did not improve. Whilst the laboratory investigated what had happened the account manager arranged for some complimentary oil analysis training as a thank you for bearing with us whilst we investigated what had happened. The clients agreed turnaround times were 2 working days for their suites and the samples in those batches had only taken on average 1.2 days to process. After much digging the process was identified as follows:
So, in this case the lab delivering a next day service was perceived to be over 1 week to end client. Following the on-site training day this was explained to the plant manager and the following solution was put in place.
There are countless other examples such as all the reports going back to one individual rather than a group so when they are off on holiday nobody sees the reports. Likewise we also had an occasion where a sampler left the samples in the back of his Van for two weeks and completely forgot to send them in.
Time Saved: 6 working days
|Case Study Turnaround (2) – Let the lab do the hard work for you
In 2014 a UK OEM and Maintenance provider of standby-generators with around 80 engineers had all the samples across the country feed back into one mailbox in which 2 individuals read each report and forwarded onto the different departments / end clients and if the report was serious a quote for the repair work would be attached when sending the samples. This was on average adding 5 days onto the Total turnaround and sometimes up to 8 days when one of the two in the team were on holiday or sick.
Discussing this turnaround concern with the engineers when they mentioned they were struggling to match competitor service turnarounds due to this delay we changed the customer over to our publishing option. In this case all the normal reports went out as usual direct to the clients, but without the intervention of the anyone, they just went automatically, but the abnormal reports that required quotations were not sent automatically. Instead each day the team of two people were able to see that publishable reports were ready. The individuals would log in, read the report and could copy and paste a hyperlink into the report to take the client to their quotation and click send. This would then send the reports to all the default clients by all the specific email rules they required. Since only 7% were caution and 3% serious this meant a 90% reduction in workload, meaning eventually one of the individuals started following up the emailed quotes with telephone calls and achieved sales person of the month soon after.
Time Saved: 5 working days