**5. Conclusion**

An attempt has been made to optimize engineering attributes of plain and non-plain weave fabrics as per requirement. Soft computing is used to solve fabric geometrical model equations and relationships between useful fabric parameters such as thread spacing and crimp, fabric cover and crimp, warp and weft cover are obtained. Such relationships help in guiding the direction for moderating fabric parameters. The full potential of Peirce fabric geometrical model for plain weave has been exploited by soft computing and the same is extended for non-plain constructions. The inter-relationships between different fabric parameters for jammed structures, non jammed structures and special case in which cross threads are straight are obtained using suitable computing techniques. It is hoped that the fabric designer will be benefited by the flexibility to choose fabric parameters for achieving any end use with desired fabric properties. This information is helpful to the weavers in avoiding attempts to weave impossible constructions thus saving time and money. It also helps to anticipate difficulty of weaving and take necessary steps in warp preparations. The relationship between the cover factors in warp and weft direction is demonstrated for circular and racetrack cross-section for plain, twill, basket and satin weave. Non plain weave fabric affords further flexibility for increasing fabric mass and fabric cover. As such they enlarge scope of the fabric designer.

Soft computing can successfully provide a platform to manoeuvre crimp in warp and weft over a wide range with only three fabric parameters; yarn tex, modular length of warp and modular length of weft yarn. This has enabled solutions by interaction of crimp interchange and crimp balance equations. This exercise offers several solutions for fabric engineering by varying the above three parameters.
