Analysing turbulence-drivers during the implementation of the pearl chain production and control system (PCPPCS).

Teich, Tobias ; Szendrei, Danny ; Unger, Katja 等

1. INTRODUCTION

The PCPPCS is a current SCM strategy. Principal component of the system is the sequencing of production orders according to the sales-order flow. The potentials of that strategy are: increasing flexibility of products and processes; implementation of lean criteria and alignment of production to dynamic customer demand (Unger et al., 2008). The greatest challenge for the PCPPCS is implementing it into running processes. It is faced with many complexities from various directions and its problems are multi-disciplinary (Weyer, 2002). As many current SCM-strategies include aspects of lean manufacturing (Bayou & de Corvin, 2008), our analysis concentrates on localising and assessing turbulence-drivers among the value adding processes of an OEM. This is to ensure appropriate FMEA-activity to get the production-segments ready for sequence-stability. Therefore we distinguish between technical and organisational turbulence drivers (td's). In section 2 we give an overview about current standards of the production system, the PCPPCS and its production planning restrictions. We then describe the segmentation of production processes and the identification of turbulence drivers (section 3). Our methodology for the analysis will be presented in section 4. Our findings and conclusions for further proceeding will be described in section 5.

2. CURRENT STANDARDS

2.1 Production System

The PCPPCS is to be implemented into an OEM plant in the automotive industry. In that plant, all work on the car body, as well as assembly of parts and modules is done. The characteristics of the plant are: series production, mixed model production, multiple working stations along conveyor systems. They are identical to the features in the work of Boysen et al. (Boysen et al., 2007). Between certain working stations there are buffers for re-sequencing, rework and blocking.

2.2 The Pearl Chain PPCS

Pearl chain is a current SCM-concept of harmonising and synchronising relevant processes of production, procurement and distribution, that all focus on dynamic demand. In detail the processes of product concept, product development and change management; material procurement, warehousing; production; sales and distribution are considered. The processes have to be implemented in terms of a static flow over all members of the sn in terms of production organisation (Boysen et al., 2007). Such a static flow can also be used for overall preventive maintenance (Schulze & Mohr, 2007). Core consideration of the ppcs is the setup of a po-sequence. This sequence is planned some days (5 to 6 days) prior to the start of production and then communicated to external members of the sn (Unger et al., 2008). From this time on, no changes of the sequence are possible. This time prior to final production at the OEM's plant enables and ensures the procurement, production or assembly of parts and modules at the member's plants. Furthermore, the requested sequence of modules can be set up in logistic's processes. An enlarged planning and producing horizon for external suppliers is achieved in that way. On the day of clearance of the po-sequence, all po's are to be driven stable, i.e. under retention of the planned po-sequence, over all working stations of the production process. All relevant activities of material disposition and warehousing have to organisationally and functionally be adapted to the posequence.

[FIGURE 1 OMITTED]

2.3 Production planning and the parameters of sequencing

Many articles concerning production order sequencing can be found in operations research literature. Most of the publications deal with modeling and defining algorithms that shall ensure reliable sequencing. Reliability in many articles targets the aspects of due-time completion of processes or due-time delivery of products (Zhou et al., 2008). In the approaches of Zhou or Baker (Zhou et al., 2008; Baker & Trietsch, 2008) the urgency of satisfactory (due-time) order completion towards customers is described and emphasised. Due-time order completion is also an important content of PCPPCS. Thus, the necessity for safe algorithms and satisfactory sequencing has been growing (Teich et al., 2008). Most algorithms have some parameters for operational scheduling of order sequences, such as: total processing time, single operation process time, setup time or certain delays in common. What is missing in production control are tools to verify and trace the process stability of a po-sequence over all working stations against the initially planned sequence. At the OEM's plant the po's are planned in order to meet assembly restrictions. These are the most important restrictions, since they consider material shortages. Other restrictions are: delivery urgency, model mix ratio, order quantity (paint blocks) or trigerring delayed po's.

3. SEGMENTATION OF PRODUCTION PROCESSES

3.1 Segmentation

In order to get detailed information about process instabilities it is essential, to reduce complex processes into clearly arranged contents of crafting. Segmentation is especially important in the body and the paint shop. There, many reasons of disturbing the planned po-sequence were found in practical trials. Certain processes of establishing a defined geometry or color (according to the po-content) were analysed. Then all necessary machinery for that single process step is captured in production segments. Incoming and outbound po-sequences can be verified for parity this way and td's become obvious. This enforces IT-systems for identifying each po. This proceeding enables the identification and localisation of turbulences drivers.

3.2 Turbulence Drivers

Turbulences degrade process stability and endanger all competitive and efficiency potentials of the concept. Turbulences occur at certain working stations. So far, turbulences were verified by chronologically counting the po's when arriving and leaving a working station. We distinguish between 2 classes of Turbulence drivers: (i) technical td's:

There is a lack of stability and reliability at the machinery in the production process. The parameters of machinery have to be adjusted to the parameters of production of all possible variance. All buffers must support the po-sequence.

(ii) organisational td's:

As mentioned above, the urgency of completion/delivery is one of the most important sequencing criteria. Regarding this restriction exclusively neglects the complexity over all po's in combination. We assume some critical combinations of po-contents. Thus detecting such critical contents in predecessor-successor-combination is necessary. These pairings can be recorded and avoided in further scheduling operations.

4. METHODOLOGY

The methodical approach of sequence analysis will be accomplished statistically on the basis of a process database (db). In that db we captured the production system structure. Via interfaces to all production segments we collect data regarding: po-sequence, po-contents and po-identification, troublesome working stations. With that structure, the posequence of any period in reality can be verified with the planned sequence. Furthermore, the sequence of the same po's, being processed in a production subsystem [AS.sub.1+n] can be verified against the sequence in [AS.sub.1]. Optimal case would be two complete identical sequences, i.e. no po changed its position within the chain. First trials in the automotive industry proved to be far away from that scenario. The db furthermore enables automatic reports about certain preset workflows. The data are analysed statistically and enable ranking of td's and thus, the priorizing of countermeasures. Results of sequence scanning have to be evaluated, prioritised and documented on a periodical basis. The evaluation must include the analysis of sequence stability over the whole process as well as over its segments. The database structure allows queries regarding: number of turbulences over time, ~ in production segments, ~ under special conditions and number of repetitive turbulences. Further activity to improve working stations regarding sequence stability is applied by FMEA (failure mode and effects analysis).

5. CONCLUSIONS

At this point, a clear definition of turbulence countermeasures is not yet possible. Our approach presents an organisational and functional solution to locate turbulence drivers. That is why we present an overall procedure to initially analyse a production system including FMEA Standards. Without localising and downsizing the impacts of td's, the potentials of PCPPCS remain theoretical. The approach is to be perceived as a pilot trial to allocate the main goals of the PCPPCS of productive and organisational resources. Our future effort will include the setup of an FMEA-organisation to priorize td's, assess the efforts of countermeasures and continue the analysis. Furthermore we establish a customizable procedure, that allows production system evaluation with integrated FMEA proposals.

6. REFERENCES

Baker, K.R. & Trietsch, D. (2008). Safe scheduling: setting due dates in single-machine problems, European Journal of Operational Research, Vol. 196, No.1, pp 69-77, ISSN 0377-2217

Bayou, Me.E. & de Corvin, A. (2008). Meassuring the leanness of manufactoring systems--A case study of Ford Motor Company and General Motors, Journal of Engineering and Technology Management, Vol. 25, pp 287-304, ISSN 0923-4748

Boysen, N.; Fliedner, M. &Scholl, A. (2007). Sequencing mixed-model assembly lines: Survey, classification and model critique, European Journal of Operational Research, Vol.192, 2009, pp 349-373, ISSN 0377-2217

Schulze, A. & Mohr, T. (2007). Preventives Qualitatsmanagement, In: Business Excellence in technologieorientierten Unternehmen, pp 97-106,

Springerverlag, ISBN: 978-3-540-73880-0, Berlin

Teich, T.; Militzer, J.; Unger, K.; Gaese, T.& Winkler, S. (2008). Ein formalisierter Ansatz zur Generierung von Kundenangeboten auf Basis heuristischer Verfahren, Proceedings of Logistics and Supply Chain Management: Trends in Russia, pp 272-282, ISBN: 978-5-7422-18104, May 2008, St. Petersburg

Unger, K.; Teich, T. & Trautmann, J. (2008). Prozessubergreifende Systembetrachtung in synchronen Produktionsnetzen der Automobilindustrie, Tagungsband ZFPro'08 Herausforderungen fur PLM und SCM-Umgebungen, pp 93-101, ISBN: 978-3-938590-15-7, Zwickau, Nov. 2008, Plauen

Weyer, M. (2002). Das Produktionssteuerungskonzept Perlenkette und dessen Kennzahlensystem, Helmesverlag Karsruhe, ISBN: 978-3-9808133-5-8, Karlsruhe

Zhou, H.; Cheung, W. & Leung, L.C. (2008). Minimizing weighted tardiness of job-shop scheduling using a hybrid genetic algorithm, European Journal of Operational Research, Vol. 194, No.1 pp 637-649, ISSN 0377-221