Tight margins on both hard and soft services pose a major challenge for facility management (FM) service providers. Those providers that optimise their services – and therefore their margins – will have greater scope to secure a favourable market position. Process Mining enables you to translate analyses into specific actions. And this provides interesting opportunities to expand services and improve margins.
Process Mining is a success story initially developed in the Netherlands. The concept of Process Mining is the brainchild of Prof. Wil van der Aalst and has been further developed over the past fifteen years at Eindhoven University of Technology (TU/e) and RWTH Aachen University.
Every organisation is familiar with workflows and automation tools. Processes, however, cannot be managed conclusively with workflows, as the chain is never completely closed. Process automation developers are often willing to accept a margin of error, whereas an accountant responsible for bookkeeping does not have such liberty. If 1,000 orders are entered into the system, 1,000 orders must also come out the other side.
Close the gap between the accountant and process manager
Process Mining closes the gap between the accountant and process manager. The solution to this sounds straightforward. Processes are followed from start to finish. This will enable you to reveal where there are hitches. Main roles are set aside for timestamps and a case ID. The case ID is an object that you track within the organisation. The timestamps indicate where that object was at any given time within the process.
You could, for example, create a case ID for every car that is registered to reserve a parking space in a car park. A first timestamp is created when a parking space is registered for the car. A second timestamp when the vehicle passes under the barrier at the car park. A third once the car is parked in its parking space and a fourth when the car departs.
Using the same example, you could also register that the car reserves a parking space but does not use it. How long does it take for the reservation of this parking space to expire? When will the space become free for another car to be parked? How long was the queue at the car park when there was an unused space available? And could this have been resolved differently?
Provide immediate insight into both optimisations and the customer journey
Spaghetti diagrams are made in the process of operational service provision, among other things. A thread (so to speak) is attached to an employee who carries out a set of tasks for a job. Once the set of tasks has been completed, we see that thread as a plate of spaghetti. This plate shows which movements have been made and where optimisation is possible. Process Mining connects a thread to every conceivable process and thus provides insight into inefficiencies in the same way. Metadata – such as location, customer, type of service and parties carrying out the work – is linked to the case. This makes it possible to produce detailed analyses and therefore to observe possible optimisations.
FM service providers can use Process Mining to get a clear picture of the customer journey. Every touchpoint is registered and the subsequent workflows and processes are mapped out. This enables service providers, for example, to further optimise their response times for ad hoc services. Various measurement points can also be linked to the case ID within every other conceivable process. How compliant am I with the agreement I made with my customer? How long does it take to process the administration of an additional cleaning assignment? How does this affect customer satisfaction?
What is the essence of your services?
The secret of Process Mining lies in the preparation. Conventional automation does not take into account the start and stop times of a process. For Process Mining, however, this is essential. Process Mining requires organisations to determine the essence of their services in advance and to link this to the case ID. Timestamps can then be attached to enable measurements to be made.
The quality of the timestamp is crucial to ensure the reliability of the analysis. It is therefore also important to keep the threshold for registering the timestamp as low as possible or even to automate this with a suitable supporting operational system. A good example of this is the use of a mobile app for employees that monitors, either automatically or semi-automatically, the start and end times of journeys. Moreover, you will also benefit from the work of TU/e, where data scientists have developed algorithms that monitor the quality of timestamps. In practical terms, this means that once the measurement has been set up, your software ensures that the results are reliable. You can easily perform your analyses from the management side.
The potential of Process Mining is really the icing on the cake. The software translates analyses from an academic tool into specific actions. As an organisation, you can determine whether you manage these actions centrally (quota assignment), enable those directly responsible to do so (show car park operators how many spaces are available) or even completely automate the process (a robot raises and lowers the barrier).
Possibilities for expanding the service provision
Process Mining applied to a system like Planon Universe for Service Providers is reliable and fast. A proper data history structure provides opportunities for long and short-term analyses. Data from the previous hour can help you take action in the next hour. That provides possibilities for optimising the implementation of the services, but also for expanding the range of services offered.
The biggest disadvantage to Process Mining is the age of the technology. Utilising an academic analysis tool in the form of a practical business tool is something quite new and innovative. That said, this ‘disadvantage’ would appear to be only advantageous. The use of Process Mining is currently gaining momentum and this provides opportunities for early adopters. Those looking for a tool that can remove the sharp edges of tight margins would be wise not to ignore what Process Mining has to offer.