Abstract
Performance evaluation models are used by companies to design, adapt, manage and control their production systems. In the literature, most of the effort has been dedicated to the development of efficient methodologies to estimate the first moment performance measures of production systems, such as the expected production rate, the buffer levels and the mean completion time. However, there is industrial evidence that the variability of the production output may drastically impact on the capability of managing the system operations, causing the observed system performance to be highly different from what expected. This paper presents a general methodology to analyze the variability of the output of unreliable single machines and small-scale multi-stage production systems modeled as General Markovian structure. The generality of the approach allows modeling and studying performance measures such as the variance of the cumulated output and the variance of the inter-departure time under many system configurations within a unique framework. The proposed method is based on the characterization of the autocorrelation structure of the system output. The impact of different system parameters on the output variability is investigated and characterized. Moreover, managerial actions that allow reducing the output variability are identified. The computational complexity of the method is studied on an extensive set of computer experiments. Finally, the limits of this approach while studying long multi-stage production lines are highlighted.
Similar content being viewed by others
References
Angius A, Horvath A, Colledani M (2011) Moments of cumulated output and completion time of unreliable general markovian machines. In: Proceeding of the 18th World Congress of the International Federation of Automatic Control (IFAC)
Betterton C, Silver S (2012) Detecting bottlenecks in serial production lines—a focus on interdeparture time variance. Int J Prod Res 50(15):4158–4174. doi:10.1080/00207543.2011.596847
Bremaud P (1999) Markov chains: Gibbs fields, Monte Carlo simulation, and queues. Springer, USA
Carrascosa M (1995) Variance of the output in a deterministic two-machine line. Master’s thesis, Massachusetts Institute of Technology, Cambridge
Chen CT, Yuan J (2004) Transient throughput analysis for a series type system of machines in terms of alternating renewal processes. Eur J Oper Res 155(1):178–197. doi:10.1016/S0377-2217(02)00838-X
Ciprut P, Hongler MO, Salama Y (2000) Fluctuations of the production output of transfer lines. J Intell Manuf 11:183–189. doi:10.1023/A:1008942917166
Colledani M, Matta A, Tolio T (2008) Analysis of the production variability in manufacturing lines. ASME Conf Proc 48357:381–390. doi:10.1115/ESDA2008-59408
Colledani M, Ekvall M, Lundholm T, Moriggi P, Polato A, Tolio T (2010a) Analytical methods to support continuous improvements at scania. Int J Prod Res 48(7):1913–1945. doi:10.1080/00207540802538039
Colledani M, Matta A, Tolio T (2010) Analysis of the production variability in multi-stage manufacturing systems. CIRP Ann Manuf Technol 59(1):449–452. doi:10.1016/j.cirp.2010.03.142
Cox D, Miller D (1977) The theory of stochastic processes. Science Paperbacks. Wiley, New York
Dincer C, Deler B (2000) On the distribution of throughput of transfer lines. J Oper Res Soc 51(10): 1170–1178
Gershwin SB (1993) Variance of the output of a tandem production system. In: Onvural RD, Akyildiz IF (eds) Proceedings of the second international conference on queuing networks with finite capacity
Gershwin SB (1994) Manufacturing systems engineering. PTR Prentice Hall, Englewood Cliffs
Grassman WK (1993) Means and variances in markov reward systems. In: Meyer CD, Plemmons RJ (eds) Linear algebra, Markov chains and queueing models. The IMA Volumes in Mathematics and Its Applications, vol 48. Springer, New York, pp 193–204
He XF, Wu S, Li QL (2007) Production variability of production lines. Int J Prod Econ 107(1):78–87. doi:10.1016/j.ijpe.2006.05.014 (special section on building core-competence through operational excellence)
Hendricks KB (1992) The output processes of serial production lines of exponential machines with finite buffers. Oper Res 40(6):1139–1147
Hendricks KB, McClain JO (1993) The output processes of serial production lines of general machines with finite buffers. Manage Sci 39(10):1194–1201
Kalir A, Sarin S (2009) A method for reducing inter-departure time variability in serial production lines. Int J Prod Econ 120(2):340–347
Li J, Meerkov S (2000) Production variability in manufacturing systems: Bernoulli reliability case. Ann Oper Res 93(1):299–324. doi:10.1023/A:1018928007956
Li J, Meerkov S (2009) Production systems engineering. Springer, Berlin
Manitz M, Tempelmeier H (2010) The variance of inter-departure times of the output of an assembly line with finite buffers, converging flow of material, and general service times. OR Spectrum, pp 1–19. doi:10.1007/s00291-010-0216-1
Meerkov S, Shimkin N, Zhang L (2010) Transient behavior of two-machine geometric production lines. IEEE Trans Autom Control 55(2):453–458. doi:10.1109/TAC.2009.2036328
Miltenburg GJ (1987) Variance of the number of units produced on a transfer line with buffer inventories during a period of length. Naval Res Logist 34(6):811–822. doi:10.1002/1520-6750(198712)34:6811:AID-NAV32203406063.0.CO;2-Z
Ou J, Gershwin SB (1989) The variance of the lead time of a two machine transfer line with a finite buffer. Techical report LMP-90-028, Laboratory for Manufacturing and Productivity, MIT
Sabuncuoglu I, Erel E (2002) Analysis of assembly systems for interdeparture time variability and throughput. IIE Trans 34(1):23–40. doi:10.1080/07408170208928847
Stewart WJ (2009) Probability, Markov chains, queues and simulation: the mathematical basis of performance modeling, 1st edn. Princeton University Press, Princeton
Tan B (1997) Variance of the throughput of an n-station production line with no intermediate buffers and time dependent failures. Eur J Oper Res 101(3):560–576. doi:10.1016/S0377-2217(96)00191-9
Tan B (1998a) Agile manufacturing and management of variability. Int Trans Oper Res 5:375–388. doi:10.1016/S0969-6016(98)00024-0
Tan B (1998b) An analytical formula for variance of output from a series-parallel production system with no interstation buffers and time-dependent failures. Math Comput Model 27(6):95–112. doi:10.1016/S0895-7177(98)00031-4
Tan B (1998) Effects of variability on the due-time performance of a continuous materials flow production system in series. Int J Prod Econ 54(1):87–100. doi:10.1016/S0925-5273(97)00132-1
Tan B (1999) Variance of the output as a function of time: production line dynamics. Eur J Oper Res 117(3):470–484. doi:10.1016/S0377-2217(98)00266-5
Tan B (2000) Asymptotic variance rate of the output in production lines with finite buffers. Ann Oper Res 93(1):385–403. doi:10.1023/A:1018992327521
Tan B (2002) State-space modeling and analysis of pull-controlled production systems. In: Gershwin S, Dallery Y, Papadopoulos C, Smith JM (eds) Analysis and modeling of manufacturing systems. Operations Research Management Science, KAP, pp 363-398
Tolio T, Matta A, Gershwin S (2002) Analysis of two-machine lines with multiple failure modes. IIE Trans 34:51–62. doi:10.1023/A:1019293215459
Tolio T, Ceglarek D, ElMaraghy H, Fischer A, Hu S, Laperrire L, Newman S, Vncza J (2010) Species—co-evolution of products, processes and production systems. CIRP Ann Manuf Technol 59(2):672–693. doi:10.1016/j.cirp.2010.05.008
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix A: Proof of Theorem 1
According to Eqs. (2) and (3):
Rearranging the previous expression
where
The sums in the second and the third terms of Eq. (37) can be expressed as known geometric sums:
By adding and removing the term \({\varvec{\pi }}{\varvec{\mu }}_\mathrm{diag} \left( \sum \nolimits _{k= 1}^{t} \mathbf{A }^k\right) {\varvec{\mu }}\), where \(A\) is a square (\(s\) \(\times \) \(s\)) matrix with identical rows formed by the transpose of the steady-state probability vector \({\varvec{\pi }}\), the following can be obtained:
It is worth to recall that, due to the general solution of discrete time Markov chains, \(\mathbf{P }\mathbf{A }=\mathbf{A }\) and \(\mathbf{A }^{k}=\mathbf{A }\) for each value of \(k>0\). Using the known sum results it is possible to write
Similarly,
Moreover, by adding and removing the term \({\varvec{\pi }}{\varvec{\mu }}_\mathrm{diag} \left( \sum _{k= 1}^{t} \mathbf A ^k\right) {\varvec{\mu }}\), the first term can be expressed as
therefore, using the known geometric sum
and, for the second term
by substituting Eqs. (44), (45) and (43) into Eq. (42) the following can be obtained:
Finally, by substituting Eqs. (41) and (46) into Eq. (37) we obtain
This is a closed-form expression of the variance of the cumulated number of parts produced at time \(t\) by a general Markovian system with transition probability matrix \(\mathbf{P }\) and reward vector \({\varvec{\mu }}\). It can be rewritten in the following form:
where:
and:
It can be easily shown that, since \(\mathbf{P }^{t}\) approaches \(\mathbf{A }\) as \(t\) tends to infinity, the term \(\frac{\beta (t)}{t}\) tails off, and the asymptotic variance rate expression becomes
The Fundamental Matrix \(\mathbf{Z }\) of a discrete time Markov chain with transition probability matrix \(P\) can be expressed as
The properties of the fundamental matrix are such that
Therefore, more compact expressions of \(\alpha \) and \(\beta \) can be obtained:
and
Therefore, the final expression of the variance is
Appendix B: Proof of Theorem 2
By referring to the partitions of the transition probability matrix P defined in Sect. 3, the mean inter departure time can be obtained with the following equation:
The first term in brackets reflects the situation in which the inter-departure time assumes value \(1\) since the system makes a transition from one operational state to another operational state or it stays in the same operational state. The second term in brackets reflects the situation in which the system makes a transition to a non-operational state and the inter-departure time increases by one unit for any time step it remains in the down state, until the system goes back to an operational state. The multiplying factor is the conditional probability of the system being in any specific operational state.
Using the results for known sums of geometric series, the following can be written:
Therefore,
It can be easily proved that, for any system, the mean inter-departure time is the inverse of the throughput, i.e.,
The variance of the inter-departure time can be expressed as a function of the second and the first moments:
The second moment of the inter-departure time can be expressed as follows:
Using the results for known sums of geometric series, the following can be written:
Therefore, the second moment of the inter-departure time can be expressed as
By rearranging the previous equation we obtain
Therefore, it is possible to express the second moment as a function of the first moment:
By substituting Eqs. (59) and (65) into Eq. (60) the closed-form expression for the variance of the inter-departure time for any general system can be obtained as follows:
Appendix C: Implementation issues
In this section, useful rearrangements of the proposed equations that positively contribute to reduce the computational time of the proposed approaches are proposed, both for the approximate formula and the exact formulas proposed in this paper.
1.1 C.1 Approximate formula for the variance of the cumulated production
Equation (5) can be rearranged in order to obtain a recursive function to be evaluated for \(k=1,\ldots ,K^*\).
where
In this way, \(C_{k}\) is a row vector that is computed by recursion at each significant time step \(k=1,\ldots ,K^*\), thus avoiding several computations of the power matrix. For a similar approach see Tan (2002). \(\rho _\mathrm{total}(\epsilon )\) then becomes
1.2 C.2 Exact formula for the variance of the cumulated production
From a computational point of view, the most complex aspect of Eqs. (13) and (14) is the calculation of the elements of the fundamental matrix \(Z\). A computationally efficient method to address this problem is described in the following. Equation (13) is rewritten in the following terms:
where \(Q\) is a row vector of the form:
where \(q\) is a row vector of \(s\) elements, easily obtained as
The vector \(Q\) can be thus obtained by solving the following system of equations:
where \(u\) is a column vector with \(s\) elements equal to \(1\). Similarly, for \(\beta (t)\)
where \(W\) is a row vector of the form
and
where \(w\) is a row vector of \(s\) elements, easily obtained as
The vector \(W_1\) can be thus obtained by solving the following system of equations:
Finally, the vector \(W\) can be obtained by solving this second system of equations:
Therefore, the variance formula reduces to
1.3 C.3 Exact formula for the variance of the inter-departure time
The same procedure can be adopted for increasing the computational efficiency in the calculation of the term \((\mathbf{I }-\mathbf{P }_{D,D})^{-3}\) in Eq. 66.
Rights and permissions
About this article
Cite this article
Assaf, R., Colledani, M. & Matta, A. Analytical evaluation of the output variability in production systems with general Markovian structure. OR Spectrum 36, 799–835 (2014). https://doi.org/10.1007/s00291-013-0343-6
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00291-013-0343-6