The decision of which job to run on which machine in which order — wrong sequence means missed delivery dates and wasted setup time (known in the literature as job-shop scheduling).
In plain words
Sound familiar?
- The morning meeting with the foreman about 'which job on which machine today' runs over an hour
- Setup (tool/fixture change) takes 30–90 minutes; on the wrong sequence it doubles within one shift
- You pay a late-delivery penalty or take a contractual score hit 1–3 times a month
- The plan lives in a spreadsheet; one machine breakdown means 1–2 hours of replanning
- Information is lost between shifts — the same job is started twice or forgotten
- Your ERP has a planning module, but it uses fixed-lot logic and does not optimize sequence
- You supply a Tier-1 automotive OEM and have to land on their audit/scorecard system
Why it matters
How it's solved
Technical depth
How it's solved
Technical depthOne-liner: Run jobs of the same material back-to-back (changeover drops), but never skip a job whose due date is closing in — the late-delivery penalty outweighs the changeover savings. The software’s job is to put a number on that gut feeling.
What the software is really doing is this: it takes the same sequencing decision your foreman makes in their head for 30 jobs, and makes it in seconds for 200. Three stages:
1. It collects the information. Which machine can do what; how long each part takes; how much changeover is needed when switching from aluminum to steel; which order is due on which day. The data is pulled from your ERP automatically, or entered once into a structured table — after that, the flow is automatic.
2. It finds the best order. The software does not try every possible sequence one by one — for 200 jobs that is mathematically impossible (the combinatorial count is astronomical). Instead, it uses optimization — a technique from operations research (a discipline that uses math and computing to solve business-decision problems) that takes intelligent shortcuts. The same answer arrives in under a minute: which job goes on which machine in which order, written out as a plan on paper. The plan jointly optimizes two objectives at once — minimum missed deliveries and minimum wasted setup time.
3. It lands on the operator’s tablet in the morning. The plan shows up as a sequence: “07:30 start job X, 09:15 switch to tool Y, 10:40 start order Z…” If a machine breaks down, the foreman does not have to rebuild the entire plan — the software recomputes only the affected part and a new plan appears in 5–10 seconds. Same for a rush order: which jobs to bring forward, which customer’s delivery can slip — shown numerically.
In short, it does not replace the foreman’s judgement; think of it as an error-free calculator that scales what they already do for 30 jobs up to 200. The decision still belongs to the operator, but they always have a current, comparable plan in hand.
Alternatives
Spreadsheet + foreman's head
FreeFree
Who it fits: 1–3 machines, 10–30 jobs/month
- + Zero cost
- + Fully flexible — the foreman can adjust on the fly
- + No capex decision
- − A sequence-dependent setup matrix cannot live in a spreadsheet
- − Plan quality degrades past ~100 jobs
- − Knowledge lives in one head — risk when the foreman is out
- − No Gantt-style visualization of machine timelines
Mid-market ERP scheduling module
Enterprise$600–2,000/year add-on on top of the ERP license
Who it fits: 10–50 employees, already running a mid-market ERP
- + Native integration with the existing ERP
- + Single-vendor support
- + Local-language interface in most markets
- − Most use fixed-lot logic — sequence-dependent setup is not supported
- − Weak 'what-if' scenario comparison
- − Module flexibility is usually narrow
Advanced planning suite (mature APS product)
Enterprise€20,000–100,000 license + €5,000–20,000/year maintenance
Who it fits: 30–200 employees, often audited by an automotive or aerospace customer
- + Mature: setup matrix, multi-constraint, scenario comparison fully supported
- + Integrates with manufacturing execution systems (MES)
- + Algorithms hardened over many years
- − High license and consulting cost
- − Rollout takes 3–6 months
- − Local-language support can be limited
Custom build on an open-source solver
Open SourceLicense free; 8–16 weeks of internal development, or $50,000–200,000 of consulting
Who it fits: SMB with an in-house software team or strong technology partner
- + No license cost
- + Fully customizable to your shop's specifics
- + Can run in the cloud or on your own server
- − Requires internal technical capacity
- − Ongoing maintenance — every solver update needs attention
- − High risk for a team without optimization experience
Recommendation
Ask in the meeting
- Do you support a sequence-dependent setup matrix? How many materials/tools can it hold?
- Can we model constraints like 'this machine runs morning shift only' or 'this machine is offline when operator Y is away'?
- When a machine breaks down, how fast does the plan update? How does the new plan reach the shop floor — tablet, ERP, phone?
- What's needed to pull data from our existing ERP — spreadsheet, direct connector, manual entry?
- Where do calendar constraints (holidays, shift patterns) come from — does your system read them, or do we enter them?
- Are late-delivery penalties factored into the objective? Can a critical customer's job be prioritized?
- How do you structure the pilot — how many machines, how many weeks, what is the success bar?
- If we stop working with you, how do we get our data back? Is there a standard export format?
Technical details
Editor’s note
This problem is known on the shop floor as “job sequencing”, “machine programming”, or simply “shop scheduling”. The academic name is Job-Shop Scheduling with Sequence-Dependent Setups (JSSP-SDST). If you go into a software demo without knowing the two terms refer to the same thing, you cannot tell whether the “advanced planning module” being sold to you actually supports a real sequence-dependent setup matrix.
The point most often overlooked in this segment: many products advertise “scheduling” but underneath only model fixed lot sizes. Information like “aluminum-to-steel changeover is 75 minutes, steel-to-steel is 20 minutes” (the sequence-dependent setup matrix) is what a real planning engine must consume — without it, finding the best order is mathematically impossible. In any demo, ask the vendor to walk through a small 5×5 setup matrix example and show you how their engine uses it.
A step-by-step path for an SMB
Stage 1 — Measure first, plan later. For at least four weeks, log three things in a spreadsheet:
- Hours spent on tool/fixture changes on each machine
- Number of late deliveries and their cause
- Idle-machine hours
Without this baseline you cannot tell which software will deliver which result, or compute payback.
Stage 2 — Build the setup matrix. Put your jobs (or job families) along both axes; in each cell write the changeover time when switching from row → column. Five to ten different materials is enough — this matrix is your knowledge capital. Any serious vendor will ask for it first.
Stage 3 — Pilot. Start with 1–2 machines for 8–12 weeks. Define the success criterion in writing, before the pilot: e.g., “late-delivery rate must drop by 50% within 60 days.” If the bar is missed, the pilot ends — that exit right belongs in the contract.
Stage 4 — Rollout. If the pilot lands, scale to the whole shop floor over 3–6 months. Operator training runs 2–4 weeks; assign one “champion” operator per shift.
Risks — what can go wrong
- Bad data. The software is only as good as the inputs. If you enter “roughly 5 minutes” for a tool change, you will get a roughly-right plan. Set up measurement before you set up software.
- Operator resistance. A foreman who does not want to use a tablet can quietly sabotage the plan. Find 1–2 volunteers in the pilot phase and put them in charge of the new process.
- ERP integration time. Vendors quote one week; in practice it is four to eight. Budget the pilot with a 2x buffer on integration.
- Vendor lock-in. Software that stores your setup matrix in a proprietary format makes it hard to migrate to a different system years later. Put a clause in the contract: “We can export our data in standard open formats (CSV or similar) at any time, on request.”
Related cautionary lesson (will be linked once published): “An 18-month ERP integration in a mid-sized CNC shop — what went wrong.”
A technical view of the solution method
This section holds what you will need when talking to a software team or a consultant. It is not what the operator sees on their tablet — it is the engine behind the curtain.
The main approaches used for JSSP-SDST:
| Approach | Typical scale | Solve time | Guarantees optimum? |
|---|---|---|---|
| MIP (mixed integer programming) | 30–200 jobs | 1–10 minutes | Yes, given enough time |
| CP (constraint programming) | 50–500 jobs | 30–300 seconds | Yes, with modern CP solvers |
| Metaheuristic (tabu, genetic) | 500–5,000 jobs | 10–60 seconds | No (near-optimal) |
| Rolling horizon | Continuous flow | Instant | Practically acceptable |
In practice: under 200 jobs, an open-source constraint-programming solver is enough. Above 500 jobs, or when you need near-instant replanning, a commercial MIP solver is typically preferred.
Objective function choice changes the shape of the solution:
- Makespan (total time): “When does everything finish?” — fits a shop running continuous work for a single large customer
- Total weighted tardiness: “Minimize late-delivery penalty” — fits a shop serving many customers with different priorities
- Total setup time: “Maximize shop utilization” — fits a capacity-constrained shop
Most real deployments use a weighted blend of all three.
Academic references
Listed in the sources block of this page’s frontmatter. For graduate-level theses, INFORMS Interfaces and the European Journal of Operational Research archive carry deployment case studies tied to real industries.
Sources
- Pinedo, M. (2016). Scheduling: Theory, Algorithms, and Systems (5th ed.). Springer. The standard reference text in the shop-floor scheduling field.
- Allahverdi, A. et al. (2008). A survey of scheduling problems with setup times or costs. European Journal of Operational Research, Vol. 187.
- Brucker, P. (2007). Scheduling Algorithms (5th ed.). Springer. The reference for algorithmic approaches.
- INFORMS Interfaces — case studies of operations research deployments in manufacturing. informs.org/Publications/Interfaces
Glossary
Related problems
A New Part Order Arrives — How Do I Get the Optimal Operation Sequence + Machine Choice + Setup?
If you are an SMB CNC manufacturer, tooling shop or engineering workshop producing 50-500 different parts, every new order puts the same decision in front of you: you take the customer's CAD model and have to work out on which machine, in which order, with which tool and fixture, and in how many setups the part will be made. If you simply write down the first feasible sequence that comes to mind, setup time grows 3-5x and parts that miss tolerance are reworked; finding the right sequence means comparing several alternative routings for the same part. When the decision lives only in one engineer's head, the similar-part memory walks out of the door when that engineer leaves, and routings of older parts are not refreshed when a new machine is bought. This page is for production-engineering teams who want to make the operation sequence and the machine assignment written and comparable as the bridge from design to make.
Backward from the Customer Order — Which Raw Material, When, and How Much Should I Order?
For mid-size manufacturing SMBs running 50-500 end items and 200-2,000 raw materials and sub-components (automotive Tier-2, white-goods component shops, furniture and assembly, machinery). With each product's parts list (BOM) and each part's procurement or production lead time on file, the question is: a customer orders 200 units of product X with a fixed delivery week — which raw materials in what quantity must be ordered which week, and which sub-parts must enter assembly when? Material requirements planning (academic name MRP) answers this question backwards from the delivery date and is the core logic underneath every ERP. Done on paper or by 'my supplier always takes 3 weeks' rules of thumb, you either run out and stop the line, or burn cash on excess stock.
Five Suppliers Quote the Same Part — How Much Do I Buy from Each?
You run a manufacturer with 20-200 staff and buy the same raw material or component from 4-8 different suppliers. Each supplier differs on price, quality, lead time, capacity, payment terms, and financial stability; deciding who stays on the approved list and how much of each order goes to whom are two separate calls. A wrong pick can trigger a months-long quality crisis or stop the line when a single supplier fails; the 'cheapest bid wins' approach hides quality, late-delivery, and compliance costs and ends up 15-40% more expensive. Scoring 5-15 bids across 8-12 criteria by hand in a spreadsheet becomes inconsistent fast.