Key figures can be an important and useful management tool in production - at least provided that they are recorded promptly and calculated on the same data basis. However, the status quo still looks different: Companies often note production data on cards and routing slips and later - if things go well at the end of the shift - transfer them to an Excel spreadsheet. This method of recording may still be sufficient for some key figures today, but it will no longer be enough to control a modern smart factory in the future.

The question is therefore: how can companies achieve intelligent, highly automated KPI recording? First of all, you should ask yourself which key figures are necessary. What are the objectives and what is the area of application? It is also important to determine your own status quo in advance. For example, companies do not need to talk about overall equipment effectiveness (OEE) if they have not yet created a basis for recording machine statuses.

The next step is to lay the foundations. A stable machine link for technical signals (MDE) and terminals for manual requalifications (BDE) are required. The data quality of the master data and shift models, which should be imported from the ERP via an interface, is also crucial. Another basic requirement is to train your employees and validate the data. This is because employees must be able to rely absolutely on the key figures, otherwise the key figure system as a whole will be questioned and not accepted. Once a KPI system has been implemented, it is essential that the KPIs are used systematically and do not 'gather dust' in a digital folder. It is therefore important to implement continuous improvement processes (CIP) in the organization.

Implementing CIP

The continuous improvement process (CIP) method consists of five phases and aims to strengthen a company's competitiveness in small steps.

© Trebing & Himstedt

A CIP typically consists of the following five phases: Plan, Execute, Measure, Analyze and Improve. In relation to the production process, the orders and production times must be planned accordingly, implemented via work instructions and recorded via terminals and machine integration. In the next phase, 'measuring', the results such as parts produced and also downtimes and faults are recorded. These results must then be analyzed and evaluated. For this purpose, the machine utilization (degree of utilization), efficiency and quality can be calculated from the measured values, from which the overall equipment effectiveness (OEE) can be derived. Prioritized improvement measures must then be derived from the key figures and implemented.

Depending on the objective, KPIs in production can initially be monetary or non-monetary and then cover the areas of efficiency, productivity and profitability. Economic efficiency and profitability indicators in production can only be implemented if a link can also be established to turnover, revenue and profit.

Step by step

But how does the data from production get into the KPI system as automatically as possible? Using the example of overall equipment efficiency (OEE), the process is divided into five steps.

The path to fully automated KPI calculation.

© Trebing & Himstedt

1. the first step is manual recording and evaluation in Excel. According to a non-representative survey conducted by Trebing + Himstedt in a recent webinar, almost 60% of companies have this status. So the question is, how can I now gradually digitize and automate?

2 This is done in step two by connecting machines with an interface for machine data acquisition (MDA). This can be a simple I/O bus terminal that supplies a simple electrical signal or - in the case of newer machines - a standardized interface such as OPC UA. This at least allows the machine status to be monitored. From an organizational point of view, the machine statuses must be defined precisely and uniformly (department/location/company-wide) for each machine type. In other words: Does the signal 'no power' actually mean an unplanned machine standstill? This logic should be conclusive and consequently less prone to errors.

3. in step three, the measured machine downtimes are qualified via the production data acquisition (PDA). If the machine does not provide an error code, an operator must carry out the requalification at the terminal promptly. It is advisable to limit the selection of downtime reasons to the minimum required in order to be able to draw conclusions. Too detailed a selection does not usually lead to better results. This provides all the data needed to determine the degree of utilization.

4 In the fourth step, an order reference can now be established via an ERP connection. The efficiency for the OEE can thus be determined using the order-related target times and cycle rates as well as the stored shift models.

5 The quality factor is still missing in order to be able to calculate a complete OEE. To do this, step five requires a good/bad counter to be integrated into the process, usually with the help of another terminal as a PDA interface. Determining the quality factor is not entirely trivial. It requires the definition of the position of the quantity count, the consideration of the influence of rework and the influence of delayed quality evaluation. If several lines or even cross-national locations are to be compared with each other, it is important to use these definitions uniformly in order to work with the same database. Key figures recorded in this way provide a convenient basis for better investment decisions, increased efficiency in production and higher customer satisfaction.

Based on its many years of project experience, the SAP partner company Trebing + Himstedt has developed the 'SAP MII KPI add-on', an SAP module that provides the areas of monitoring, KPI calculation and evaluation in a standard package - including corresponding standard reports and optimized screen designs.

The right tool

After a short implementation phase, the starter package determines the most important reasons for downtime over a period of three months. Together with the process experts from Trebing + Himstedt, further recommendations for action are then developed on the basis of the measured data, leading to higher capacity utilization and improved delivery reliability.

At the heart of the package is a preconfigured server with all the necessary SAP manufacturing licenses and the SAP MES module from Trebing + Himstedt, the SAP MII KPI add-on and a mobile tablet for simple and intuitive recording of the reasons for downtime on the store floor.