Interfacing Vending machine with the Internet.

[Wharty]

Hi, I am new to the site and looking for some advice/help.

I have Zero experience of the Pi and have no programming experience so I am not even sure if what I am about to ask is even possible with the raspberry Pi or if this is the right section of the forum to ask.

If what I am about to ask is possible then I am willing to buy the Pi and learn to program.

I would like to develop a Barcode based controller to interface with a vending machine.

It's a proof of concept that I am looking to get together and I believe Pi may be able to do this.

I basically want people to be able to order their lunch on their mobile phone and then collect that lunch from a vending machine at their place of work.

Once they have placed their order they will be sent a Barcode. They then scan the Barcode at the machine and the machine then presents the lunch to the customer.

That's the easiest way for me to explain and I am sure there is much more to it than that but i would be grateful if anybody could give me some advice.

Thanks, Lee.

[jamesh]

That sounds perfectly feasible.

[FLYFISH TECHNOLOGIES]

Hi,

Nice product idea...

What sounds a bit weird is that you're willing to buy RaSPi and learn programming only if this idea is feasible (and it is, as Jamesh already responded). My first reaction (and advice you're asking for) is that there is probably quite some hurdles to jump over (not to forget time spent of doing this) before you'll compose a production system. Consequently, your motivation is to be much stronger than it radiates from this your message...

Few additional suggestions:
- use QR codes instead of barcodes,
- you might add another identification options (NFC comes to mind) and from marketing perspective you could have also "loyalty cards" used for identification, which would be printed RFID cards. This technology is cheap...
- make properly modulated system architecture (that you can easily add new identification system, for example),
- do not forget to have online real-time monitoring with various alarm and warning triggers set - leaving somebody without his/her lunch would not help the business growing,
- if I were you, I'd implement also remote control feature, where you would be able to remotely let the person standing next to the machine to collect his/her lunch after the intervention (mobile phone can stay out of power or can be left somewhere, etc.); create an emergency exit for such cases.
- ... I need to stop thinking further... else you'd get scared.


Best wishes, Ivan Zilic.

[Wharty]

That sounds perfectly feasible.

Thanks Jamesh at least that's a start.

[Wharty]

Hi,

Nice product idea...

What sounds a bit weird is that you're willing to buy RaSPi and learn programming only if this idea is feasible (and it is, as Jamesh already responded). My first reaction (and advice you're asking for) is that there is probably quite some hurdles to jump over (not to forget time spent of doing this) before you'll compose a production system. Consequently, your motivation is to be much stronger than it radiates from this your message...

Few additional suggestions:
- use QR codes instead of barcodes,
- you might add another identification options (NFC comes to mind) and from marketing perspective you could have also "loyalty cards" used for identification, which would be printed RFID cards. This technology is cheap...
- make properly modulated system architecture (that you can easily add new identification system, for example),
- do not forget to have online real-time monitoring with various alarm and warning triggers set - leaving somebody without his/her lunch would not help the business growing,
- if I were you, I'd implement also remote control feature, where you would be able to remotely let the person standing next to the machine to collect his/her lunch after the intervention (mobile phone can stay out of power or can be left somewhere, etc.); create an emergency exit for such cases.
- ... I need to stop thinking further... else you'd get scared.


Best wishes, Ivan Zilic.

Hi Ivan and thanks for your reply.

I don't think it's weird at all, I wouldn't want to buy a Pi only to find out I couldn't achieve my goals with it. You see I was also told about Arduino and a Gadgeteer box?? I wanted to find out which one would work so my apologies for the confusion I may have caused. I don't like wasting money and believe me I am highly motivated.

Thanks for your suggestions and please don't scare me anymore

I frequent the another forum for Businesses on a regular basis and what you have said about NFC and RFID has been mentioned but from a business point of view it would be an added expense to give each user a card. You would have to pay for the card and pay for postage. The cards would also stop impulse buyers who see the machine. You could see the machine, get an account and order your meal straight away without waiting for cards. If you don't have your phone with you then I have considered a Touchscreen where the user logs in and orders/ collects food from the machine. The screen would be paid for by not having to pay for/deliver cards.

So I guess what I need to know is, where do I start with all this?

Are their any local Pi clubs that i could join in my area where enthusiasts would be willing to help me or is this online community the place to be to ask for help and advice?

Are there people looking for projects to work/help on? I am happy to pay for all the parts and give recognition for the work done.

I had a PM and I was advised Python may be worth learning? I really don't know where to start with learning Pi or with building this machine, so any advice would be great.

Thanks.

[ame]

My advice would be to listen to what people on this forum say. Not all of them, and not because they are saying don't do it, but because a few people here are smart and can offer good advice. We also have a lot of experience (with the Pi, and hundreds of other devices) and programming skills.

There are two aspects to what you are doing. The first is how to make it work (and at the base level it seems feasible), but the second is how to break it. If someone suggests a failure mode and you can not suggest how to deal with it then you're going to be sad because things *will* break.

Anyway, the idea seems reasonable. I have one question. Is the vending machine pre-stocked with meals, or is the machine loaded with meals that are made *after* the customer places an order with their phone?

[Wharty]

My advice would be to listen to what people on this forum say. Not all of them, and not because they are saying don't do it, but because a few people here are smart and can offer good advice. We also have a lot of experience (with the Pi, and hundreds of other devices) and programming skills.

There are two aspects to what you are doing. The first is how to make it work (and at the base level it seems feasible), but the second is how to break it. If someone suggests a failure mode and you can not suggest how to deal with it then you're going to be sad because things *will* break.

Anyway, the idea seems reasonable. I have one question. Is the vending machine pre-stocked with meals, or is the machine loaded with meals that are made *after* the customer places an order with their phone?

Thanks for the advice.

The user will go online and choose their meal for the next day or the next 5 days in advance. Once we get the order we will then deliver the meal to the vending machine. I imagine the box with the meal would have a barcode on it. I imagine scanning the barcode when loading the machine and the machine remembering where the meal/barcode is. The customer comes along with their phone and the machine scans the barcode on their screen. The barcode will match the one stored in the machine for the particular meal/customer and it would then present that meal.

It's basically a "Click and Collect" vending machine for lunches.

The idea came from being unhappy with the sandwich vans that came to my old workplace. I was often left choosing something i didn't really want because the van had been to 6 or 7 other businesses first. I want to be able to guarantee that a customer gets the meal they want and i think a "Click and Collect" machine will do the job well.

[FLYFISH TECHNOLOGIES]

Hi,

I wouldn't want to buy a Pi only to find out I couldn't achieve my goals with it.

I don't want to argue with you on this one, but let's take a look to the overall picture - my very raw estimation is that this decent system requires at least 6 EM (engineering months) of implementation effort + material costs. In this context the cost of one RasPi is far from being significant...

I frequent the another forum for Businesses on a regular basis and what you have said about NFC and RFID has been mentioned but from a business point of view it would be an added expense to give each user a card. You would have to pay for the card and pay for postage. The cards would also stop impulse buyers who see the machine.

I mentioned additional ID method.
RFID cards are cheap, if you give one to a regular customer and include it with his/her 10th order, it costs you a penny for the card and nothing for the postage.

So I guess what I need to know is, where do I start with all this?

Since you mentioned a PM, I expect that (s)he made a WBS, what is one of the mandatory fundamental documents.
Go trough this WBS and estimate the effort of each item. If an item requires more than 10 ED, split it further to subtasks.
At this point you'll realize that there are some items where you can't estimate the effort. They are good candidates to be added into the risks table. Afterwards, process this risks table - take some time to think about preventive actions to avoid each identified risk. As a result, you'll get a list of most of your very next activities.


Best wishes, Ivan Zilic.

[Richard-TX]

An internet connected coke machine is nothing new.

CM had one back in the 80's.

Most of them have gone away.

Here is one page that describes one.

https://www.ucc.asn.au/services/drink.ucc