Software as a service (SaaS) is a business model for deploying an application—in this case, pipeline leak detection—via the internet. By using this deployment method, a software provider can license its products in various ways to suit the needs of its client. Typically, SaaS applications are used where the user does not have the resources to manage the applications locally, or does not wish to incur the large up-front cost of ownership.

Leak-detection capability, in one form or another, is something every pipeline must have. Pipelines have an excellent safety record, but accidents still happen.

An example of such an accident occurred in Burnaby, British Columbia, Canada, in July 2007. A construction crew was performing a sewer excavation when they hit an oil pipeline. Despite normal precautions in planning at the work site, a mistake was made and the oil pipeline was ruptured. The oil geyser spewed for more than 30 minutes after the initial damage. This is a typical leak scenario where some event ruptures a pipeline, causing an unexpected economic loss and environmental damage. Having a leak-detection system in place allows an operator to know within minutes if there is a leak and to take swift appropriate action to minimize the damage.

Currently, the only available leak-detection systems are purchased applications that are installed on-site. These tend to have very expensive initial costs, installation and maintenance. A SaaS leak-detection implementation works by having a SCADA system exchange data with a remote application. The results are returned to the SCADA system and are also available to other users as needed. A SaaS-based leak-detection application would be more affordable to a pipeline operator who does not want the initial expense of a traditional system, but still desires leak-detection capability.

Implementing a realistic leak-detection service is certainly within the limits of the current technology. The question becomes: Is it something to be considered? Depending on available connectivity, not all computational pipeline-monitoring (CPM) methods may work well. Methods that are time sensitive or data intense may not perform well with a slow internet connection. Whereas, simpler methods, such as compensated mass balance, might work just fine for a single pipeline using a dial-up connection.

Once the user has a clear understanding of what to expect, further information can help managers make an informed decision on the type of system needed. Specifically, five key areas of information are invaluable when determining whether or not to choose a service-based leak-detection monitoring system: Cost, data security, communications, performance, and support. Also, as in every management decision, there are two sides to consider: Pros and cons.

The pros
The initial cost and personnel requirements are the primary reasons for using a SaaS application. SaaS products are normally billed on a monthly basis. As such, the cost is usually treated as an ordinary operating expense. This may be extremely important to an existing pipeline operator when capital-budget funds may be scarce. Another advantage is the ability to easily add or delete functions as required to manage costs. With today’s expanding business use of the internet, the costs for communication equipment and service options should continue to decrease over time.

Because SaaS is an off-site application, the data is kept in a more secure environment. Typically, the SaaS vendor will have redundant servers in multiple geographical locations. Therefore, managers gain an increased probability that their data will survive an incident at their own data center. Storing the data in an off-site application eliminates the need for multiple server locations, as well as the cost to maintain them, and protects data by providing an automatic backup of it.

Meanwhile, SaaS enables personnel to easily access data from multiple sites without having to build and manage an in-house approach. This access can be achieved by using the internet with virtual private network (VPN) connections, leased lines, or web-based access using secure HTTPS. Depending on how a data stream is tapped, using an off-site service may even allow an operator useful data when other systems are down.

But, what about performance issues? Not to worry. A SaaS application usually has a minimal performance impact on the local system because the majority of the processing is performed remotely. Just a small interface program is required for communication between the local SCADA system and the remote application. The vendor will have the ability to monitor specific data from his site and address any problems that might arise to keep the application running at its peak performance.

Also, SaaS vendors are motivated to provide excellent support to the clients on an ongoing basis. With the SaaS application, the vendor wants to make sure the bill gets paid every month. With a traditional system, while an executive is negotiating to spend $300,000 or more, he has the vendor’s best technicians’ full attention.

After the system is commissioned, these key personnel generally move on to the next client and the executive is now working with a second-tier support group. With a SaaS system, it is reasonable to expect top-shelf knowledgeable expertise to help with any issues that arise or changes to the system during the life of the contract. Having this level of expertise available frees managers from having to maintain in-house expertise or contract with an expensive third-party provider for occasional changes.

The Cons
Just as there are reasons for using a leak-detection service, there are reasons against using a SaaS-based application. Addressing the same items as before, a monthly expense must be paid or service is turned off. This is not a problem with a purchased system. In the event of budget cuts, a manager might drop the annual maintenance, but must still have a leak-detection system. By having a purchased system and not paying any monthly expense for leak-detection, management can free up funds for other activities, such as additional training or more personnel.

As for out-sourcing data, managers should consider that security, industrial spying and snoopy reporters are always an issue when dealing with data. As long as data resides “inside the fence” data is under control of the pipeline operator. Once the data leaves its facility, the operator no longer has direct control and must rely on its vendor's security measures. Security concerns now include not only its own facility, but the network connection and the vendor's facility.

Communications are another factor to consider. A reliable data link between a company-owned data center and the vendor’s can be intimidating to setup. Unlike most service applications, leak detection requires 24-hour access to the vendor. A dedicated circuit can be installed, but that is expensive and unlikely. The next most likely alternative is a VPN connection. Setting up a VPN connection may require a special router and an experienced IT specialist to correctly configure the connection.

Under normal conditions, transferring data to the vendor's machine is not a problem. When the internet begins to lag or slow down, the quality of the leak detection may degrade or even fail. The use of dial-up, satellite or cellular data can be used as a backup connection but will require additional complexity at the company’s site.

Finally, when a leak-detection system package is purchased, it belongs to the pipeline company. If a vendor drops support, the leak-detection system is still operational unless something breaks. With a SaaS application, if the supplier ceases operation or is bought by another company, service could be canceled or experience huge cost increases. Also, owning the system allows the operators to decide if and when to install updates. With a SaaS product, management may not know when updates are installed. Managers should take action to be aware of how the vendor handles updates and what happens if an update should fail.

Overall, these factors, and possibly others, should be carefully considered when undertaking the business decision of whether or not to use a service-provider leak-detection system. In the end, each company has many different moving parts and there is no “one size fits all” solution.

Leak-detection as a service appears to be a low-risk and cost-effective alternative to a traditional purchased system. The low startup cost allows for a low-risk venture into leak detection for even the smallest pipeline operators. The monthly cost is typically higher, but is counted as an ordinary operating expense verses a major capital expense. Also, a SaaS application allows a pipeline operator to install a system quicker than waiting one or more years for budget approval. In the future, if a better alternative becomes available, the pipeline operator can easily switch without having to scrap a large investment. Leak-detection service is a viable alternative to the traditional product, especially for smaller pipeline operators who do not wish to make a major investment.