I agree with Bryan that a inhouse application will give you more headache on the long run. But there are some advantages of a temporary in house system.
When we started looking for a cmms in 1993, we had a semi manual system. We had an equipment masterlist wit all our assets defined and coded. That was a simple table with the assets most relevant data.
We were recording the number of PM's and IM's (incidental or corrective maintenance) in another table based on history.

We still have magnetic boards were the PM's of all critical equipment are set out for the year, with color codes defining the type of service and the status. After all this years we still use it in combination with the CMMS. Until now I have not met the CMMS that can beat the impact of our visual planboards of 3' x 8'with the equipment year schedules devided in 52 weeks.

A training in Maintenance Planning and Scheduling was conducted by the New Standard Institute for the whole maintenance staff and some critical clients, and with the knowledge gathered we set up a 6-month pilot work order system as preparation for a demonstration of the most likely CMMS vendor.
It was simple and we brought in the culture of doing things through work orders. No fancy stuff, plain tables, look-up constraints and queries were constructed on the fly. The stock numbers used for the jobs were inserted with code and quantity. When we wanted to know the costs we just imported the list from the stores on floppy and matched them.

The stores and procurement still are responsibility of the finance directorate, but at that time, the stores supervisor had his own system developped in dbase, just like Maintenance had two paradox geeks.
We had a succesfull demonstration with lots of inputs from both maintenance and stores, and we had to wait for the next release of the software, because the current version did not support all the basic features we wanted.

Our two separate custom systems which we knew inside out made it possible to demand the best solutions.
Today paradox and dbase data are all on oracle in a CMMS that has:
Inventory management, purchasing, maintenance management, investments, budgetting, contracting, documentation and much more, and more important is linked to Oracle Financials for the Finance administration. We consider our self a medium sized company (650) employees

In setting up an access based system I would consider some basic points:

1) The maintenance part
- recording the work done
- defining workorder priorities
- defining critical and non critical equipment
- defining the equipment owners (other departments)
- defining the mrc's (maintenance resposibility centers or executers of maintenance)
- documentation
- maintenance strategies
- maintenance budgets
- levels of authorization (work, purchasing, monitoring)
- costs (how are the costs linked to the jobs)

2) The stores part
- inventory control
- stock levels
- reorder points
- direct charges
- services
- procurement (who is authorized to buy, at what level?)

3)Security
- number of users
- user rights
- hardware/software, access is a desktop database good for cookbook receipts or personal adresses, unless it is implemented by trained experts.
- financial auditing

One word of caution, CMMS are big, once you start developping, the sooner you will switch to a canned system to get the work done, because you will see your limitations.
It happened with us in a time when good independent consultants were scarce.

If your maintenance park consists only of 10 pieces of equipment, and your whole maintenance workforce has 3 people, no spare parts or stores, an inhouse Access system could work.

I am afraid that I can see no advantage in developing a temporary system. Many of today's CMMS systems allow cut and paste from other applications making implementations quite easy. For SME's there is no reason why they should not go right ahead with their implementation.

I am commenting on this as one who has tried both methods. Access is a great tool but you need highly skilled developers to get the best from it. Generally they will know nothing about plant maintenance so it is difficult putting a spec together. You tell the developer that you need something, he/she nods his/her head and comes back three days later with the wrong soution. You have to then say that this was not quite what you meant and start again. Believe me, in house applications are not easy to develop. As I have already said why would you want to do this when you buy good, well supported, shrink wrapped solutions for under a thousand bucks?

The reality is that developing in house will end up costing you many thousands.

I am going to be the oddball in this discussion. I do not entirely disagree with the other post however; I felt the some of the same animosity for Off The Shelf (OTS) CMMS’s when I started looking for a CMMS 8 years ago. To me it was like trying to drive a square peg in a round hole, I could make it work but it was not pretty. One thing I would absolutely agree with is NOT hiring or using an on staff developer to design your CMMS. You will end up with something that has no relation to Maintenance Management. On the other hand if you have a Mustang Maintenance Manager who has come up through the ranks and understands what the Maintenance Mechanic needs to do his/her job and is pretty computer savvy then you could end up with the perfect CMMS. When I first started developing mine I was up and running within a year including building the LAN. I admit that I had several iterations after that but they all built on the previous implementation. When I transferred from Puerto Rico to Florida I had the opportunity to start from scratch. I started entering baseline data in August, went on line in October, and the WO flow had leveled and was running smooth by December. For the past 2 ½ years as I manage the Work Order system I am always tweaking and/or improving it. I have a flexibility that cannot be matched by OTS systems. If I need a report I make it on the spot, just about every query is printable or I can drop in Excel. I have a system that does every thing I want it too; but I am not so foolish as to believe that this is the solution for everyone. I can honestly say it has worked for me and I cringe at the thought of implementing Maximo. I have yet to fine anyone in my organization that likes Maximo; that is anyone on the deck plates. Upper level management thinks it’s great but as is typical will not help with the implementation, they just want the results. Oh well, gotta go to work!

CMMS systems can be great tools or a nightmare. It's all in how you deploy it.

The SME's should drive how the programmer develops the database and it's features. The programmer works for the SME's....not the other way around. I have seen it this way though.

The goal is to get what you need in a CMMS system based on what your processes are designed to do. Often the CMMS is purchased then the processs are mapped. This is happening on a current project that I am involved in. The IT people are driving how and what the system does without input from the users. This is the wrong way to deploy such a system.
Tail wagging the dog syndrome once again!

I have been developing my own CMMS systems before there were CMMS systems and have been responsible for interfaces on a capital project developing a CMMS for a Govt. contractor and Access will work fine as long as you don't exceed size limitations.

I have experienced both excel and access crashing when the file sizes rose above 60MB. (This limitation may not exist with the latest version of Access.) In access you can create a query that creates a report when a certain parameter in a field is updated and have it send an automated e-mail to responsible individuals when the update occurs.

If you are going to be pulling vast amounts of data then Oracle or SQL Base or something along those lines will be required. However they are what I call pure databases and require some sort of user interface to be developed where you can see the data. The popular method is to use a Web browser to access the data. This may be done by utilizing java script for controls and VBscript or other advanced features to create a connection to the database and the desired table and fields. This is an advnced method and would require a programmer.

You can obatin a free version of Oracle from Oracle. There are also inexpensive programs that will read any ODBC compliant database, tell you the fields size and data type, read the data, convert it into another format and output it to pretty much anywhere you want it.

CWorks has/had a free version of a CMMS obtainable on the internet.

Development in house is costly and leaves you at the mercy of your current programmer. You are better of purchasing something that you can work into phases for cost. If you want i have a low costing system that i have been using and it is very inexpensive and gives you the flexability to prioritize you planning when cost is a big concern.

Jesus A. Pacheco
Jesus.Pacheco@tempowebworks.com
http://www.tempowebworks.com