Using process mining to improve productivity in make-to-stock manufacturing

Figures & data

Figure 1. Process mining procedure for productivity improvement in manufacturing.

Table 1. Format of exemplary event log for process mining.

Case ID	Activity	Timestamp	…
EX001	Moulding	2021-01-25 07:58:20h	...
EX001	Assembly	2021-01-25 08:05:03h	...
EX002	Moulding	2021-01-25 08:32:19h	...
EX001	Packaging	2021-01-25 08:45:37h	...
…	…	…	...

Table 2. Event log for process mining at Geberit.

–	Case ID	Activity	Timestamp	–
Event	Part ID	Process (Reader)	Scan time	Variant ID
1	P.0232907	Moulding (Reader $R_{1,2}$)	2019-04-13 08:00:58h	V.01
2	P.0232612	Assembly 2 (Reader $R_{3,2}$)	2019-04-13 08:00:59h	V.02
3	P.0232157	Sorting (Reader $R_{4,1}$)	2019-04-13 08:01:00h	V.18
4	P.0232913	Moulding (Reader $R_{1,1}$)	2019-04-13 08:01:05h	V.18
…	…	…	…	…

Figure 2. As-designed process model.

Figure 3. As-realised process model.

Figure 4. Percentage of total throughput time per segment.

Notes: The figure only contains complete cases. The whisker length is the 1.5 interquartile range and the median is highlighted as yellow line. Significance level: *** p < 0.001, **p < 0.01, *p < 0.05.

Figure 5. Visualisation of historical process flows at capacity constraint.