The Orion Data Network – Oil & Gas asset visibility case study. Active RFID 
The requirement is to make the supply chain transparent by automatically showing the location of items within the supply chain. Previous attempts at doing this using passive tags failed for a number of important reasons. In particular it was necessary to get within a few centimetres of the tags to read them, which presented a severe health and safety issue when dealing with the large pipes in transit. With that in mind a trial of long rage RF ID was planned. This would remove the need to be close to the items, automate the audit process and deliver the data centrally. This document describes the technical aspects of the successful trial carried out in February 2012 in the port Of Batumi, using the new, Orion network long range RFID tags. It then describes the key elements and timescales to be met, if the system is to meet the operational deadlines or in service date.

The Trial In Outline

The trial consisted of pipes being unloaded from the ship to the quayside and then onto the rolling stock. The system recorded the pipes being unloaded and audited the pipes in storage at the quay side. The system then enabled the user to identify each flat bed and account for the pipes being loaded onto it.

Each element of the trial was tested using 50 sample tags, specifically programmed for the trial’s overall requirements. The tags used a ‘blink’ architecture enhanced with motion transmission. Having conducted a brief trial at a site in Glasgow, it was already known that the tags could perform.The trial was to test the tag solution’s operating procedures and the ‘in field’ equipment required to make it as simple as possible. All the data from the field trials was stored and analysed to enhance the parameters of the protocol used. The tags take their settings from the local gateway and as such can take on different profiles for different tasks. The trial has provided the necessary data to tune the system.

Lessons Learnt

A number of important lessons were learnt on the ground. For ease of use on site, the reader application that talks to the tags should be split into two different pieces of equipment. Firstly there should be a hand held version which plugs into a windows tablet with GPRS/Wifi connectivity, as opposed to a laptop. This should have its 433MHz aerial mounted on to an extendable boom, enabling the operator to move about the site easily and take readings at will. The second piece of equipment should be an Ethernet enabled (i.e. office based) reader that can be left at sites to produce the automatic audit. Both these pieces of equipment could be supplemented with repeaters and active RFID Tags.

Essentially the problem encountered on the site during the trial was that the actual loading and unloading area was quite confined.The stack of pipes, which for health and safety reasons is only stacked three high, extends for perhaps 100 metres along the quayside. If there is only a few metres or tens of metres around the stack it is hard to place the reader or gateway device sufficiently far enough away to see them all. The solution is to simply walk passed the stack; it reads the tags in blocks. The alternative solution for fixed installs is to place repeaters around the edge of the area so that they can then relay the signal to the gateway. (For clarity ‘reader’ shall mean a handheld device and ‘gateway’ shall be a fixed install).

GPS RFID Tags

Note the short distance between the ends of the pipe and stack of containers in this image, compared to the length of the stack of pipes.

Other aspects that were looked at included the mounting of the tags in the pipes. This was done using a mounting tool that placed the tag around two metres within the pipe. The tag itself was held on by four neodymium magnets. The tag was in a case with an aerial at ninety degrees to ensure the best possible signal.

Orion Tags Oil and gas

An image to show a tag – they were placed well beyond arm’s reach within the pipes for the trial.

The batteries used on the tags are lithium 3.6v Tadiran units and these will last in the order of six months. It would be possible to go to an alkaline solution in the future but this would cut the life span, as cell for cell the lithium batteries have three times the energy density. The tags can have a replacement battery fitted during their lifetime.

Feedback

The feedback from the users was enlightening. The health and safety aspects of working in a busy port should not be under estimated. The range of the tags meant that where the ground permitted they could be fifty or more metres away from the action and out of harm’s way. However the confines of the site meant that they had to walk alongside the rail cars and pipe stack to conduct some tests. Because of the nature of the signal radiating from the pipes, particularly when placed on the flat beds it would be easier if they could mount a repeater on each flatbed. In practice the aerial on the boom would also provide a simple solution. (This is because when standing too close the signal is largely radiating above head height). Oil and Gas tracking tags

Note the height of the pipes on the carriage, the bottom of the flatbed is approximately four feet off

The range of the tags combined with their motion sensor also meant that when the testers had identified each individual rail car and assigned it on the system, they were able to withdraw some distance and pick up the tags which were being lifted. It was easy then to spot which rail car they were going to. A final walk pass confirmed the tags where correctly allocated. They found it took around four minutes for each load to be placed on the rail car, given two or three lifts per carriage and ten rail cars there was plenty of time to record the results and then double check them. With the application on a tablet and an aerial on one hand, a single user could comfortably perform all the tasks well within the time it takes to load or unload at the quayside.

Production Timescales and Tasks

The biggest task with going into production is managing the lead time on the tag components. Currently this is three months. During the trial is was mentioned that the system might have to meet an August deadline which means time is already short if the tags are to be shipped to the supplier in time to be loaded with the cargo. It will also be necessary to train the suppliers and install a gateway at their sites.

There are then a number of small development tasks to improve the case and aerial set up on the tags.This includes modifying the case to use moulded in magnets rather than screw in ones and widening the base to make the potting of the aerial easier to contain. These are minor, non-risk issues that are easily within the tag production time frame.

On the application side there is approximately fifteen man days to port the application to a suitable tablet and tidy up the user experience. No decision on the choice of tablet has been taken, other than to say that it should be a standard ‘off the shelf’ Windows product.

In parallel to the tablet work, an FTP (File Transfer Protocol) website will be put in place to collect data from units and also to enable new applications to be issued to readers. In other words the application that sits on the tablet and drives the scanner will be fully supported online.

The website will also list FAQs and contain user support documents to aid in the training and deployment phase of the project. It was noted in the trial that the equipment was easy to use but a small amount of time spent on basic RF orientation would benefit users in general. The readers and gateways are low cost, under £500. In theory a reader will plug into any USB enabled Windows device, enabling BP to make use of any pre-existing equipment or simply source tablets at the best possible price. For all the developments mentioned, it is envisaged that these will be done on receipt of a tag order and are not therefore individually costed items.

Management software

The FTP site will send information to the Management software database that will need to be tailored and built, along with a user interface, which again will be web based. The user software will also include information already supplied by suppliers such as Pipe numbers, Barcodes, Passive tags as well as certification. Other information can also be fed into the software such as VT3 tracking data, for creating routes etc. The software will allow a user to create a manifest which can include as much or as little info about the shipment as they require, and give full visibility of the supply chain in near real time. It also becomes a historic database where information can be pulled at anytime about a shipment, an individual pipe/item or the certification relating to a pipe/item or item. This software will take approximately 3 months to build, given that all relevant information from suppliers is made available quickly.

Conclusions

The trial went well and generated enough data to improve the tag protocol and define the ‘in field’ equipment that would be required by the users.

Amendments will be made to the application for the user. The scope of the trial was to test the solution, rather than deliver the final user interface. This and a small list of other features for users, such as a website for support, will be done when the solution goes into production.

The critical path to the operational solution is dominated by the time it takes to manufacture and ship tags to suppliers at the start of the logistics process. It can be achieved for the proposed August start date, so long as the project kicks off as soon as possible