為了達到對化工廠進行整體監控的目的，我們必需要能夠得到各個重要程序變數的準確估測值，因此過去有關感測器網路設計的研究多半著重測量點配置。但除了準確度以外，事實上系統可靠度也是一個須研究的實際課題，在文獻中僅Sanchez與Bagajewicz [2000]曾經探討過流量感測器網路之單純修正型維修策略，但其他維修手段(如備份及重複硬體等)卻仍然付之闕如，而本研究之主要目的即是開發兼具重複與備份硬體之感測器網路維修策略，並以改良式基因演算法搜尋對應最適化問題最佳解。我們發現備份硬體對系統可靠度會有顯著的幫助，而重複硬體的裝設對提升可靠度的效果雖不如備份硬體顯著，但卻可同時增加估測準確度。在成本較充裕的情況下，採用兼具重複及備份之維修策略可以使系統可靠度與準確度同時超越傳統感測器網路所能達到的極限值。

Accurate estimates of the key variables are considered to be essential for assessing the performance of any chemical process. In order to enhance estimation precision, one of the main themes in past researches is concerned with the optimal placement of sensors. However, other than the problem of sensor network design, the issue of maintenance policy should also be addressed to ensure the system reliability.
In the literature, only Sanchez and Bagajewicz (2000) tried to analyze the impact of integrating a simple corrective maintenance program in the design of flow sensor network. Although reasonable results have been obtained in this study, it should be noted that the discussions of other means for enhancing the reliability, e.g. preventive maintenance, redundant and spare hardware, have not been included at all. Consequently, it is the objectives of this study to incorporate the possibilities of redundant and spare hardware in a corrective maintenance model and also develop a mathematical program for generating the best sensor locations and their maintenance strategy simultaneously in a mass-flow network.
The genetic algorithm were used in our application examples to solve the optimization problem. From the results we have obtained so far, it can be concluded that
The spare sensors can indeed be used to significantly raise the system reliability.
Although adopting redundant hardware is a less effective method for improving reliability, it can be applied to reduce the estimation variability as well.
If the life-cycle cost of a sensor network is not a limiting condition, the maximum precision and reliability achieved with the conventional design and maintenance strategies can always be surpassed with the use of proper redundant and spare sensors.

Ali, Y., and Narasimhan, S.“Sensor network design for maximizing reliability of linear process”AIChE J., vol. 39, pp. 820-828, May 1993.

Ali, Y., and Narasimhan, S.“Redundant sensor network design for linear process”AIChE J., vol. 41, pp. 2237-2249, October 1995.

Ali, Y., and Narasimhan, S.“Sensor network design for maximizing reliability of bilinear process”,AIChE J., vol. 42, pp. 2563-2575, September 1996.

Bagajewicz, M. J.“Design and retrofit of sensor networks in process plants”AIChE J., vol. 43, pp. 2300-2306, September 1997.

Bagajewicz, M. J., and Sanchez, M. C.“Cost-optimal design of reliable sensor networks”Com. Chem. Eng,. Vol. 23, pp.1757-1762, 2000.

Bagajewicz, M. J. Process Plant Instrumentation : Design and Upgrade Technomic Publishing Co., Inc, Pennsylvania, 2001.

Bagajewicz, M. J., and Sanchez, M. C.“Design and upgrade of nonredundant and redundant linear sensor”AIChE J., vol.45, pp1927-1938, September 1999.

Bagajewicz, M. J., and Sanchez, M. C.“Duality of sensor network design models for parameter estimation”AIChE J., vol. 45, pp.661-664,March 1999.

Bhushan, M., and Rengaswamy, R.“Design of sensor network based on the signed directed graphy of the process for efficient fault diagnosis”Ind. Eng. Chem. Res., vol. 39 pp.999-1019, 2000

Blischke, W. R., and Murthy, D. N. P. Reliability : Modeling, Prediction, and Optimization, John Wiley & Sons, Inc., Canada, 2000.

Deo, N. Graph Theory with Application to Engineering and Computer Science, Prentice Hall, Englewood Cliffs, NJ, 1976.

Dorsey , R. E. ; Mayer , W. J. “Genetic Algorithms for Estimation Problems With Multiple Optima,Nondifferentiablility,and Other Irregular Features” Jounal of Bussiness & Economic Statistics , 13 , 1 , January , 53 ,1995

Fong, C. C., and J. A. Buzacott,“An Algorithm for Symbolic Reliability Computation with Pathsets and Cutsets”IEEE Trans. Rel., R-36,34, 1987.

Garrard , A. ; Fraga , E. S. “Mass exchange network synthesis using genetic algorithms” Comput. Chem. Eng. 22 , 12 , 1837, 1998.

Goldberg , D. E. “Genetic and Evolutionary Algorithms Come of Age” Communication of ACM , 37(3) , 113, 1994.

Henley, E. J., and Kumamoto, H. Probabilistic Risk Assessment : Reliability Engineering, Design, and Analysis, IEEE Press, New York, 1992.

Michalewicz, Z. Genetic Algorithm + Data Structure = Evolution Programs-Third, Revised and Extended Edition, Springer, New York, 1996.

Madron, F., and Veverka, V.“Optimal selection of measuring points in complex plants by linear models”AIChE J., vol. 38, pp.227-236, February 1992.

Mah, R. S. H., Stanley, G. M., and Downing, D. M.“Reconciliation and rectification of process flow and inventory data”Ind. Eng. Chem. Process Des. Dev., vol. 15, pp.175, 1976.

Narasimhan, S., and Jordache, C. Data Reconciliation & Gross Error Detection : an intelligent use of process data, Gulf Publishing Company, Texas, 2000.

Raghuraj, R., Bhushan, M., and Rengaswamy, R.“Locating sensors in complex chemical based on fault diagnostic observability criteria”AIChE J., vol.45, pp.310-322, February 1999.

Schaffer , J. D. ; Caruana , R. A. ; Eshelman , L. J. ; Das , R. “A Study of control Parameters Affecting Online Performance of Genetic for Fuction Optimization” The Third International Conference on Genetic Algorithms and Their Applications, 51, 1989.

Sanchez, M. C., and Bagajewicz, M. J.“On the impact of corrective maintenance in the design of sensor networks”Ind. Eng. Chem. Res., vol. 39, pp.977-981,2000

Sen, S., Narasimhan, S., and Deb, K.“Sensor network design of linear process using genetic algorithm”Com. Chem. Eng, vol. 22, no.3, pp.385-390, 1998.

Yeh , C. H. “A Literature Review on the Components Improvement of Genetic Algorithms” Journal of Taiwan Water Conservancy , 44 , 1 , March ,1996.