The literature says there are both cubic and hexagonal phases in my system. The peaks that I have manually indexed are both showing different planes from what the literature has suggested.
Since you obviously know the phases and the lattice parameters why you don't insert both in simple equations to calculate the d-values and then the reflection position using Braggs equation?
Gert Nolze, I am not sure if there is only one phase or two. And I do not know the lattice parameters too. Vegard's law would give me a close approximation for the lattice parameters but not the correct values. All I have is the theta(angle) and the data on the elements used for making the alloy
Nevertheless, there are simple options to use the theta values, calculate the d values and estimate which hkl fits to these values. Especially if you already have a rougth estimation about the lattice parameters. You can also use PowderCell and insert there the approximate lattice parameters (and load the XRD data as x_y-file (containing theta and intensity. Then you can simply play a bit with the lattice parameters. If it should be a hcp structure use 194 as space group number (the atom at 0.3333 0.6667, 0.25). For the cubic phase either 225 for fcc or 229 for bcc (in both cases one atom at 0,0,0). In case of more complicated intermetallic phases is becomes anyway more problematic since these structures may produce superstructure reflections which might be very small, or the structure is complex. However, in the last case you wouldn't ask for cubic and hexagonal since the you would see more peaks.
PowderCell is also able to refine the lattice parameters if your assumption (model for the lattice parameters) is already close to the peak maxima, i.e. only slightly shifted.
There are several programs for indexing the powder diffraction pattern - Conograph (http://research.kek.jp/people/rtomi/ConographGUI/web_page.html), Crysfire (http://www.ccp14.ac.uk/tutorial/crys/), Dicvol, which is part of Winplotr software (http://www.cdifx.univ-rennes1.fr/winplotr/winplotr.htm). There are many others (ITO, TREOR or N-TREOR, X-Cell, ...).
Dear Balasubramanya , Fullprof is another technic that provides you to give your parameters as inputs. Theoretically Mr. Nolze 's opinion is true. But in practice, it is not easy for everyone to make XRD pattern fits using programs like FullProf, Maud, GSAS etc.