As pointed out in the comments, there is no way to make it safe, as methyl vinyl ketone itself is a very nasty compound.
I am also not an expert on organic synthesis, but I found some procedures, that might help you in adopting a synthetic route. The one you suggested might work, but be aware, that you are dealing with a methyl ketone, which can undergo the haloform reaction quite easily.
For example, in  they describe the bromination with bromine and absolute methanol at the methyl group for 3-methyl-2-butanone at a nice moderate temperature range 0-10 °C. The yield is 95%. But this is not what you want. Adding an acid as catalyst might work in your favour though, as after protonation the partial positive charge is better stabilised at the higher substituted carbon.
In  they use N-bromosuccinimide and aqueous hydrogen bromide to introduce the halogen in α position. They convert hexanoic acid with thionylchlorid to the acid chloride and subsequently to 2-bromohaxanoylchloride. The reaction takes place in the also quite nasty carbon tetrachloride at about 85 °C. This might also work for your system, it's a bit warmer than you expected though.
In  they use sulfuryl chloride in carbon tetrachloride to convert pantane-3-one to 2-chloropentane-3-one at about 45 °C with about 80% yield. How selective this will work with an asymmetric ketone is not clear.
Once you have obtained the 3-halogeno-butan-2-one, things will not get easier. I found a patent , where they convert 3-chlorobutan-2-one via pyrolysis with sodium acetate at 500-550 °C.
I could not find a reference for an elimination, but that should also work as long as you use a bulky base.
Wikipedia  has a different route for the synthesis, which might be a bit more economical and easier.
Well, good luck!
- Org. Synth. 1976, 55, 24.
- Org. Synth. 1976, 55, 27.
- Org. Synth. 2002, 79, 251.
- J. R. Long: Preparation of methyl vinyl ketone. US 2256149 A, 16. 09. 1941.