In　the　experimental　study　on　hydrogen　storage　properties　of　solid　materials　like　metal　hydride,　the　estimation　of　the　kinetic　parameters　in　the　model,　which　is　often　done　by　least　squares　data　fitting,　assumes　great　importance.　Unfortunately,　the　parameters　thus　obtained　for　the　same　material　even　show　great　disparities　in　the　literature　reports.　Besides　some　known　factors　like　non-isothermal　operation,　these　disparities　may　also　be　attributed　to　the　poor　selection　of　the　design　variables　of　the　experiments　for　data　fitting.　In　this　paper,　we　present　some　theoretical　tools　following　the　idea　of　optimal　experimental　design　(OED)　to　analyze　the　error　in　the　estimated　parameters,　which　is　caused　by　the　error　in　the　experimental　data　and　related　to　the　design　variables.　We　found　from　the　analysis　results　that　after　optimization　of　the　design　variables,　the　confidence　on　the　estimated　parameters　can　be　effectively　improved.　Copyright　©　(2014)　by　the　Committee　of　WHEC2014　All　rights　reserved.