Eurotunnel ISIS/ICIS 1994 - 2015 Reflection - Final

Editor's Notes

1. A project originally conceived in 1994 following the opening of the Channel Tunnel and ‘completed’ in 2005. The Integrated Customer Information System.
2. I joined Eurotunnel as a Consultant in 1992. Sutton, London. Moved to Farthingloe Village in 1993 Moved to Cheriton Parc in 1994, End of the Construction Phase Beginning of the Production/Exploitation Phase
3. The Project centres upon mechanisms for communicating operation events from the 4 Control Centres THIS is the UK control centre in Folkestone 1994, Following the Opening, It was soon realised that Poor Consideration had been given to the Provision of Information. Designed as a Shuttle Service, Turn Up and Go. (Like the Tube). Despite Highly Technical Systems implemented for.......... Zero means were available of communicating the operational Status Quo The system was (Probably) designed in 1980s before Internet & Web Applications were Ubiquitous. Here are technical systems, Fire Alarms, Tunnel Equipment Monitoring Train Route/Path Controlled by a Rail Traffic Management Computer/Operator All Communication from within the control centre, externally is Voice Phone/Radio Based. What needs to be Communication ? Staff - Platform Changes ? Customers - Waiting Time No Delivery Mechanism for Operational Information No Delivery Mechanism for Customer Information KEY TO UNDERSTANDING THE VALUE OF EACH DEVELOPMENT & ITS PRIORITY WAS TO ANALYZE EACH BUSINESS PROCESS & THE IMPACT OF EVERY OPERATIONAL EVENT
4. 1994 I developed a Pilot System using Simple Forms (BOFP Framework) (Back of Fag Packet) Very Successful ! Six Staff were Employed as “Information Assistants” to man the post 24/7 They were Sited Directly next to the RCC Controller. Conveyed Operational Events directly via the simple character cell forms. Depart, Arrive, Cancel, Delay, Change Platform
5. As with many ‘Pilot Systems, It became THE System Demands Quickly Grew beyond Depart & Arrive Application started to Grow Organically. Following Several years of Successful Operation........... Requirements Continued to grow. Often Faster than we could deliver............. We used Primitive Business Process Modelling As the Freight Service Grew, MC4 Function was developed As TMAR Service grew, Dangerous Goods.... As Tourist Service Grew, Tolls Function was Developed As WWW & Reservations grew, Toll Interface & Reservations Interface was developed to predict waiting times for Passengers. Single Deck Management was Implemented.
6. Following sdeveral years of Successful operation, in 1999 It was decided to Redevelop the application. Redevelopment Sought to Adopt the use of Modern Process Modelling Tools Adopt modern Development Tools & User interfaces, (Visual Studio) Implement Multi-Tier Client Server Technology Adopt OO methods to (promote Re-Use of Code Components) Ease maintenance There was Technology Change The Business was Rapidly Changing, Refining Processes & Resolving Early Start-up Problems.
7. Rational Rose Use Case Modelling This Collaboration diagram shows Class Interaction and Event Sequence
8. THIS is what we were aiming at......... THIS is the Product that is still used to this day......... By Looking at the Screen an instant view of the Operational Status Quo can be gained. For example.............. 6 are Tourist, 7 are Freight, 4489 are Through-Freight (No DG) Tourist 18/34 has come from Yard Traffic State is Normal On Freight, 53 Vehicles on-site. Toll Queue on M20 not terribly busy Early versions used DCOM. Problems of Scalability, Blocking. Re-engineered to use MSMQ.
9. THIS is how we worked to document requirements, Plan This was Pre-ITIL but Each Change was Documented, Described, Submitted for Approval, Sought weekly Pre-Agile but Domain Experts were represented in the core Project Team, Requirements were Prioritised, Costed , Developed, Tested & Implemented as Phased Releases Much Like Sprints in Latter day SCRUM.
10. 2003 asked to extend ISIS to include all customer facing signage Previous Focus was Staff, Now to Include Customers Good Engineering Practice was paying off, More Sophisticated Use was more-easily achieved. Single Deck Planning ICIS live in early 2005 I left in 2005. Same system is in use today.
11. Why was it so Successful. We were Highly Motivated, We were Automating many Manual Processes All were Able to see the business benefit of our efforts Quickly & See the Direct Impact Upon the Service. For example, the VMS Signs had been in place for >10 years but only displayed fixed messages ‘Have a nice day’ We Brought them to life by exploiting their capabilities, By displaying relevant information.
12.  Dynamic systems development method
13. A story seldom told One of which I am immensely Proud.