Page 295 -
P. 295

Chapter 11
            Analyzing “Lasagna Processes”

















            Lasagna processes are relatively structured and the cases flowing through such pro-
            cesses are handled in a controlled manner. Therefore, it is possible to apply all of
            the process mining techniques presented in the preceding chapters. This chapter
            characterizes Lasagna processes and discusses typical use cases for process mining.
            Moreover, the different stages of a process mining project for improving a Lasagna
            process are described. The resulting life-cycle model guides users of process mining
            tools like ProM. Moreover, different application scenarios are discussed.



            11.1 Characterization of “Lasagna Processes”


            Unlike Spaghetti processes, Lasagna processes have a clear structure and most cases
            are handled in a prearranged manner. There are relatively few exceptions and stake-
            holders have a reasonable understanding of the flow of work. It is impossible to
            define a formal requirement characterizing Lasagna processes. As a rule of thumb
            we use the following informal criterion: a process is a Lasagna process if with lim-
            ited efforts it is possible to create an agreed-upon process model that has a fitness
            of at least 0.8, i.e., more than 80% of the events happen as planned and stakeholders
            confirm the validity of the model. This implies (assuming that a suitable event log
            can be extracted) that all of the process mining techniques presented in this book
            can be applied.
              The spectrum ranging from Lasagna processes to Spaghetti processes is a con-
            tinuum. Sometimes the terms “structured”, “semi-structured”, and “unstructured”
            are used to refer to the same continuum. In a structured process (i.e., Lasagna pro-
            cess) all activities are repeatable and have a well defined input and output. In highly
            structured processes most activities can, in principle, be automated. In semistruc-
            tured processes the information requirements of activities are known and it is pos-
            sible to sketch the procedures followed. However, some activities require human
            judgment and people can deviate depending on taste or the characteristics of the
            case being handled. In unstructured processes (i.e., Spaghetti process) it is diffi-
            cult to define pre- and post-conditions for activities. These processes are driven by
            W.M.P. van der Aalst, Process Mining,                           277
            DOI 10.1007/978-3-642-19345-3_11, © Springer-Verlag Berlin Heidelberg 2011
   290   291   292   293   294   295   296   297   298   299   300