I sat through a few presentations to an airport customer looking to purchase a PSIM. It was a learning experience finding out the separate needs and wants of the various departments who would be the system managers and end users.
The single largest detractor was cost.
Some products focused their pitch heavily on how it could bring multiple systems together and allowed changing of key components later without creating issues with the users. As an example, this airport through the years had two primary access control products installed in quantity and at least three VMS solutions to integrate. As the airport migrated to a single product of each the dependency on integrations would fade somewhat. Some products focused their pitch on how an emergency with a strong work flow and following reporting requirements could be easily managed across required inputs such as elevator, fire, escalator, gates, intrusion, people flow, access control, first responder access and more. Some focused extensively on the reduction of lifecycle training for operators by having a workflow driven common platform. Shorter training cycles and more useful and skilled personnel.
I guess until you have sat in an emergency center dispatch chair with radios and alarm going on a critical event and then had the biggest boss ask for something "NOW" that required pulling data from three systems it's hard to place a value on these things. With that said......they are expensive to install and maintain. A benefit is tying in 10 different systems. That becomes an albatross when you can't/won't/aren't ready to update one system and it holds you back on others or it breaks the system etc.
Great article. You have touched on a few of the major limitations of PSIM. Here are a few more:
Most system configurations and settings still have to be performed using the application software of the underlying systems, not through the PSIM. Most PSIMs give you an integrated interface on the monitoring side, but not on the configuration side. This is a huge disappointment for end-users who think that they are getting a truly unified system.
All underlying applications need to continue to be supported and upgraded. Most upgrades will require that the PSIM "driver" that connects to the application also be upgraded, sometimes costing tens of thousands of dollars per upgrade.
The feature set of the combined systems usually defaults to the least capable feature set of all of the systems being integrated. For example, if one access control system supports global anti-passback, and another one doesn't, chances are good that the overall system won't have global anti-passback.
I believe that Steve Hunt, creator of the Security Dreamer blog (not very active these days) claims credit for coming up with the "PSIM" acronym. We have had a few lively debates over the years as to the future of PSIM. He had a greatly expanded view of how PSIM could be used, arguing that the data generated from such a system could be used not just for security, but for a multitude of other business related purposes.
I am sure there is someone out there that thinks that but I doubt it's a common approach. Mostly because case management software is far less expensive than PSIM so that's a lot of extra money to spend if you do not really need all of it. Also, I'd be surprised if most PSIMs have the depth of case management that real case management software offerings deliver.
"No end user objects to the idea of PSIM .... It is the price and time investment"
The question then becomes - how can you get the cost down? For that, you need to make integration quicker while increasing scale (i.e., the number of customers using / buying it). Of course, that brings us back to the core problems - the market is fragmented, making scale very hard and integrations are typically time consuming and limited.
Most of our (bigger) customers want, buy and use our PSIM solutions. With over 400 interfaces to the most common manufacturers in the building technologies field, i do not see any problems - only positive aspects IMO. If you have 10+ subsystems with a whole lot of components and need them to interact based on various scenarios, you can not rely on simple I/Os or such. Furthermore, since many companies are outsourcing security personel, you need a common GUI everybody understands without the use of 10+ different trainings or GUIs.
In the end you have to be flexible - no interface yet? well we or our partner develop it to our customers needs. Function not included? same here.
..a circumstance, maybe none of the smaller integrators are able to or capable of and therefore neglect PSIMs.
"i do not see any problems - only positive aspects IMO"
The problem is cost for everyone but your 'big' customers. For sure, someone who has 10+ subsystems has spent millions or tens of millions on systems already, so they likely can justify spending a million more to add PSIM. Yes/no?
I agree with you here, John. I would only add that even some of the "big" customers are recognizing the long term cost aspects of multi-tiered complex integration models versus minimizing the number of integrations where possible.
The above is simply one example. There are obviously various iterations of multi-tiered integrations using many types of systems depending on the End-User customer.
John brings up a good point about huge investments in lots of systems where some EUers need to develop easier methods to "monitor". I certainly agree that the PSIM models have there place with a very select type of customer. However, they have to recognize the long term affects of their decision as the cost of the back-end support and administration may actually be much more than spending the money to invest in a product that could potentially minimize the number and levels of integration required.
This article brings back so many memories. Back to the days when we were trying to figure out what we were going to do about PSIM. Our parent business unit at the time saw it as a strategic imperative that they be able to leverage what PSIM seemed to offer. They were interested in the biggest integration projects and felt the need to offer everything to everything functionality. And not just about data integration but also workflow, ‘rules engines,’ geospatial, lots of reporting. PSIM to them was another integration platform arrow in their quiver. Pelco’s interests were a little more tactical--we just needed to stave off competition from other VMS that had more/better integration with third party access or intrusion systems. There was a tricky balance to be had accomodating use cases in our own VMS vs giving up others to the PSIM. In both cases we were looking at the upper right quadrant on a chart with criticality in one axis and integration complexity on the other axis.
What happened next might be predictable. Look at all the possibilities, get overwhelmed by the permutations, admit that no one product would solve all needs, decide to ‘partner’ with all of them. Basically, we went from a build-vs-buy-a-PSIM discussion to deciding to do nothing.
As a software abstractionist it didn’t surprise me. Successful integration frameworks emerge when many integrations happen and common patterns emerge that can be encapsulated into reusable components. The best integration frameworks (that is, those that actually work), are the result of many many iterations. PSIM would never quite get there for a number of reasons: lack of standards, lack of sufficient economic drivers, simply lack of enough jobs that required the integrations. PSIM was really about taking snapshots of integration implementations and trying to say those were products. Without enough customers in need of exactly that same implementation, there’s no market.
“PSIM is not the right term for where we’re heading; PSIM is the wrong terminology for the right technology,” he says. “Until the industry comes up with a better term, there will continue to be confusion about what PSIM is and isn’t. I wouldn’t be surprised if a new term emerged within the next six to 12 months.”
We currently use a PSIM type software in Immix by Sureview to do video monitoring of our customers facilities. They had a failed attempt to move us to their new Cloud version which didnt perform as well as the Enterprise version, which they stopped any new development on. We use is for video monitoring and customer incident retrieval from the website so not much access integrated into it. Are there any other platforms comparable?
I like this article as it touches on a lot of key points that I've been dealing with over the last 4-5 years. And I agree that the word PSIM is thrown around loosely without thought. That beings said, we've successfully deployed a PSIM platform and is currently integrated and running, connected to 27 different sub-systems from CCTV, ACS to Elevators, Escalators and even controlling lock-boxes for tender documents.
The product we're using is probably the most expensive, but is of them all the most expandable and customize-able one out there. IPSecurityCenter by CNL is the product we're using and we also create our own in-house driver base to connect to almost ANYTHING.
The KEY thing in doing so is proper project and sub-system management. This touches on something mentioned earlier around "Version Management". At first this was our biggest headache as sub-system vendors would come on site maintaining their equipment and upgrading software willy-nilly causing driver incompatibility. However we caught this issue early on and could establish corporate policy for change-control to avoid this.
Running and maintaining it is a full-time job and requires a lot of continuous communications with the end-user to provide the information (the "I" in PSIM) that he/she NEEDS, not wants. A PSIM platform that does not change during its lifetime is now a big white elephant. If your PSIM platform doesn't allow for these changes, you've invested in the wrong product.
The biggest benefits we've achieved with our deployment are: > Less operators to train to monitor the vast amount of sub-systems > Expose sub-system maintenance flaws > Speed up maintenance procedures by passing info directly into ERP Plant Maintenance Applications (like SAP) > Expose lazy operators > Manage security staff > Minimize false alarms (consolidating different sub-system events as a single true alarm rather than creating alarm instances for each sub-system event) > Many more
In the end it's how you drive it and how willing you are to manage and maintain it.
#8, that's great to hear and thanks for your first comment!
we've successfully deployed a PSIM platform and is currently integrated and running, connected to 27 different sub-systems from CCTV, ACS to Elevators, Escalators and even controlling lock-boxes for tender documents.
That's a lot of equipment and certainly, a scenario where DIY or trying to hack things through a VMS or EAC system would be infeasible.