I teach the chemical engineering process design course at Yale University and in this course I used the following textbook:
Peters, M. S., K. D. Timmerhaus, and R. E. West, “Plant Design and Economics for Chemical Engineers,” 5th Edition, McGraw-Hill, ISBN #0072392665, 2003
This textbook has a chapter on equipment cost estimation and one of the methods used in Marshall-Swift (M-S) indexes methodology.
To use the M-S index, you divide the index for the date for which a cost of equipment is desired by the index for the date of the known cost and multiplying the resulting factor by the known cost.
The M-S Equipment Cost Indexes are based on a national average for 47 different industries. A cost index is made up of a market basket of equipment and indicates the general direction in which equipment costs are moving. They represent an estimate of the trends in installed equipment costs from 1914 to date.
Yehia Khalil mentioned a great textbook that I used myself extensively throughout my M.S. but can also be coupled with other texts such as Turton, Bailie, Whiting, and Shaeiwitz Analysis, Synthesis, and Design of Chemical Processes which comes with a cost estimation tool (CAPCOST) that can help a lot for cost estimation.
As for the M & S Index, depending on your target cost estimation precision, you may use Dr. Kiran Patil's suggestion or create a linear regression of the historical indices and predict the 2016 index from there. Capital costs tend not to be a majority of the annualized cost but you can always do a sensitivity analysis to understand how sensitive your economic output is to the chosen model parameters
Cost estimation comes with great uncertainty (Breaking the Model: Why Most Assessments of Biobased Materials and Chemicals Costs Are Wrong; Mark Bunger)
and depending on your level of analysis, take appropriate measures.